Umeå University Medical Dissertations, New Series No. 1362

PARASTOMAL HERNIA

Clinical Studies on

Definitions and Prevention

Arthur Jänes

Umeå University 901 87 Umeå Sweden www.surgery.umu.se 2010

Department of Surgical and Perioperative Sciences

2

Tutor..........................…....Leif A Israelsson

Financial support...…........FoU-Västernorrland

Responsible publisher under Swedish law: the Dean of the Medical Faculty

This work is protected by the Swedish Copyright Legislation (Act 1960:729)

ISBN: 978-91-7459-052-4

ISSN: 0346-6612

E-version available at http:// umu.diva-portal.org/

Printed by: Arkitektkopia

Umeå, Sweden, 2010

3

Inget är så dåligt att det inte kan bli sämre av en operation Lars Wikström, Hudiksvall

To Joel, Jakob, Simon and Elin

4

5

TABLE OF CONTENTS

ABSTRACT 6

ARTICLES INCLUDED 9

INTRODUCTION AND BACKGROUND 10

AIMS OF THE STUDIES 22

PATIENTS AND METHODS 23

RESULTS

Article I + II 28

Article III 31

Article IV 35

DISCUSSION 38

CONCLUSIONS 47

REFERENCES 48

ACKNOWLEDGEMENTS 58

APPENDIX - THE ARTICLES

Article I

Article II

Article III

Article IV

6

ABSTRACT

Parastomal hernia is a frequent complication after stoma formation occurring in up to

80 % after five years. In previous studies, no consistent definition of parastomal

hernia has been used at clinical examination or at CT-scan.

Aims of the studies

To evaluate the short and long term effects of prophylactic prosthetic mesh placed in

a sublay position at the index operation on the development of parastomal hernia and

stoma complications. The purpose was also to validate a definition of parastomal

hernia at clinical examination and a method and a definition of parastomal hernia at

CT-scan.

Patients and Methods

A. 27 patients were randomized to a conventional stoma and 27 patients to the

addition of a prophylactic low weight, large pore mesh in a sublay position.

Follow up was after one month, one year and five years.

B. In routine surgery 93 consecutive patients were provided with a

sigmoidostomy with the intention of receiving a prophylactic mesh in a sublay

position. Follow-up was after one year.

C. 27 patients with stomas were examined by three surgeons and parastomal

hernia was defined as any protrusion in the vicinity of the stoma. CT-scans

were assessed by three radiologists with patients examined in the supine and

prone positions. Herniation was defined as any intra abdominal content

protruding beyond peritoneum or the presence of a hernia sac.

Results

A. Without a prophylactic mesh parastomal hernia was present after one year in 13

of 26 patients (50%) and after five years in 17 of 21 (80%). With a mesh

parastomal hernia was present after one year in 1 of 21 patients (5%) and after

five years in 2 of 15 (14%). There were no mesh related complications.

7

B. A prophylactic mesh was used in 75 of 93 patients. In 9 a mesh could not be

utilized due to scarring after previous surgery. In 9 a mesh was omitted because

of the surgeons’ decision. A mesh was used in 19 severely contaminated wounds.

Wound infection occurred in 6 of 73 (8%) patients with a mesh and in 4 of 15

(27%) where no mesh was used. Parastomal hernia was present in 8 of 61 (13%)

patients with a mesh and in 8 of 12 (67%) where no mesh was used.

C. 27 patients were assessed by three surgeons and three radiologists. Kappa was

0.85 for the surgeons and 0.85 for the radiologists, with CT-scan in the prone

position. Kappa was 0.80 for surgeons and radiologists collectively, with CT-scan

in the prone position. Four parastomal hernias detected with CT-scan in the prone

position could not be detected in the supine position. A clinically diagnosed

parastomal hernia was always detected with CT-scan in the prone position.

Conclusions

A prophylactic mesh placed in a sublay position at the index operation reduces the

rate of parastomal hernia without increasing stoma complications. A mesh can be

placed in severely contaminated wounds. Parastomal hernia should at clinical

examination be defined as any protrusion in the vicinity of the stoma with the patient

straining in a supine and an erect position. At CT-scan, with the patient examined in

the prone position, herniation should be defined as any intra abdominal content

protruding beyond peritoneum or the presence of a hernia sac.

Key words: Parastomal hernia, stoma complications, prosthetic mesh, mesh complications,

sigmoid stoma, ileostomy, laparoscopic stoma, CT-scan

8

This picture shows a large parastomal hernia.

9

ARTICLES INCLUDED

This thesis is based on the following articles

I. Arthur Jänes, Yucel Cengiz, Leif A Israelsson

Preventing parastomal hernia with a prosthetic mesh

Archives of Surgery 2004 ; 139 (12) : 1356 - 1358

II. Arthur Jänes, Yucel Cengiz, Leif A Israelsson

Preventing parastomal hernia with a prosthetic mesh : a 5-year follow-

up of a randomized study

World Journal of Surgery 2009 ; 33 (1) : 118 - 121

III. Arthur Jänes, Yucel Cengiz, Leif A Israelsson

Experiences with a prophylactic mesh in 93 consecutive ostomies

World Journal of Surgery 2010 ; 34 : 1637 - 1640

IV Arthur Jänes, Lena Weisby, Leif A Israelsson

Parastomal hernia : Clinical and Radiological Definitions

Submitted

10

INTRODUCTION AND BACKGROUND

There are historic reports of stomas occurring spontaneously after trauma or

incarcerated hernias developing spontaneous fistulas. The latter seems

sometimes to have saved the patient’s life for a period of time - but most

certainly with a very much diminished quality of life. One such instance was

Ms. Margreth White treated in 1740 by Mr. Cheselden for a spontaneous

ostomy from an incarcerated umbilical hernia. Surgeons also surgically

opened incarcerated groin hernias with subsequent ileostomies developing.

In such instances spontaneous closure sometimes seems to have occurred.

Olof Acrel, in 1752 surgeon at Serafimerlasarettet, Stockholm, Sweden,

presents such a case in his surgical memoirs “Chirurgiske Händelser”

(Inspärradt tarmbråck i högra ljumsken, upbrutet af tilslagen brand och läkt

utan fistel). In 1710 Littré suggested colostomy as a surgically useful

procedure. In 1779 Pillore performed the first caecostomy and in 1793 a

colostomy was performed by Duret on an infant with anal atresia. The first

successful colostomy with long term survival may have been performed by

Pring in 1819.

Ms. Margreth White was in 1740

treated by Mr. Cheselden for an

incarcerated umbilical hernia that

formed a spontaneous ostomy.

11

Today creating ostomies surgically is a common procedure in both emergent

and elective surgery. It is a procedure employed in infants with congenital

anomalies as well as in adolescents and elderly with acquired maladies or

functional disorders. An ostomy may be intended to be permanent but is also

used solving emergent situations, as protection of an intestinal anastomosis

or as preparatory of definitive surgery with the aim of restoring bowel

continuity at a later stage. An ostomy can be created during open as well as

laparoscopic surgery. If bowel is resected or cut the bowel end is brought out

as an end ostomy. Especially in emergent situations or in situations when the

ostomy is not judged to be needed for a long period of time the full bowel is

exteriorized and a loop stoma is created. The development of ostomies as a

frequent procedure has been enabled not only by the progress in methods of

anesthesia and surgery but also by the development of modern stoma

bandages. The terms ostomy and stoma (Greek: mouth) are often used

interchangeably but have somewhat different meanings. An ostomy refers to

any surgically created opening in the body for the discharge of body wastes.

A stoma is the actual end of a hollow viscus that protrudes through the

abdominal wall. In the present context focus is on ostomies consisting of

ileum or colon brought out through the abdominal wall by a surgical

procedure.

In 1989 Pearl regarded parastomal hernia as an incisional hernia

related to a abdominal wall stoma [1]. Parastomal hernia is a common

complication and Goligher even considered some degree of parastomal

hernia to be an almost inevitable complication of colostomy formation [2].

Several attempts have been made with different surgical techniques when

creating an ostomy in order to lessen the rate of parastomal hernias

developing but with discouraging results [3]. Once a parastomal hernia has

developed it is very difficult to repair and an abundance of methods have

12

been tried. Such repair has often been associated with poor outcome and

high recurrence rates [3-4]. A parastomal hernia often causes the patient

discomfort and problems with stoma bandages including leakage. In 2008

Kald showed that patients with parastomal hernia, defined as bulging at the

stoma site, had a significantly impaired quality of life [5]. On the other hand a

well functioning ostomy without herniation may allow a very good quality of

life [5-9].

Definition of Parastomal Hernia

Currently the generally accepted definition of an inguinal hernia is a

protrusion in the groin when the patient is straining in the supine and the

erect position [10]. Similarly an incisional hernia is defined as any protrusion

in the vicinity of the incision when the patient is straining in the supine and

the erect position [11]. There has, however, not been any generally accepted

definition of parastomal hernia. A parastomal hernia may be regarded as an

incisional hernia related to an abdominal wall stoma [1]. Parastomal hernia

has been classified into four subtypes: the subcutaneous type with a

subcutaneous hernia sac, the interstitial type with a hernia sac within the

muscle / aponeurotic layers of the abdomen, the perstomal type with the

bowel prolapsing through a circumferential hernia sac enclosing the stoma,

and the intrastomal type in ileostomies with a hernia sac between the

intestinal wall and the everted intestinal layer [12]. In clinical examination it is

difficult to distinguish between these four subtypes of parastomal hernia so it

has not been used in clinical studies.

In surgical reports over the years there has not been any uniform

definition of parastomal hernia used at follow-up examination [3-4]. The lack

of a uniform definition is a major problem when assessing and trying to

compare rates of parastomal hernia presented in different clinical reports or

13

trials. Until 2003 the definition of parastomal hernia used at follow-up was

actually given in only one report. Then herniation was defined as a palpable

cough impulse at the ostomy site [13]. In all other clinical reports before 2004

no definition of parastomal hernia was given at all [4] . In the years following

2004 several authors have defined parastomal hernia as a palpable defect or

bulge adjacent to the stoma [5, 14-21]. In two studies a CT-scan was added

to the clinical examination and the radiological definition of a parastomal

hernia was any abdominal content protruding along the ostomy [14, 22]. The

correlation between parastomal hernia detected at clinical examination by

several examiners has not been very good [23]. The correlation between

herniation detected at clinical examination and at CT-scan has been rather

poor. Herniation present at clinical examination may not be detected at CT-

scan and vice versa [22-24]. This is a problem that needs to be solved and a

consistent definition at clinical examination that correlates to findings at CT-

scan should be sought. Agreement between assessments is often analysed

by calculating the Kappa value. Kappa has a maximum of 1.0 when

agreement is perfect. A value of zero indicates no agreement better than

chance and negative values indicates agreement worse than chance. The

strength of agreement is often interpreted as Kappa 0.6-0.8 as good and 0.8

or higher as very good.

Some authors have differentiated between parastomal hernia and

stoma prolapse [25-31]. In these reports neither the definition of parastomal

hernia or stoma prolapse was given so it is unclear how they differentiated

between these conditions at clinical examination. In a Cochrane report in

2007 on loop stomas a difference was made between stoma prolapse and

parastomal hernia [32]. Prolapse was defined as an eversion of the stoma

through the abdominal wall and parastomal hernia as a hernia beside the

stoma. The authors did not give any guidelines on how to differentiate

14

between these two conditions at clinical examination. Many authors may

have considered both entities as a parastomal hernia.

The lack of a uniform definition and the almost total lack of any

definition presented in reports before 2004 makes it difficult to compare the

rates of parastomal hernia between different studies. It is generally agreed

that follow-up should be no less than 12 months after the index operation in

order to detect a ventral incisional hernia and this seems to be applicable for

parastomal hernias also [33]. However, the rate of parastomal hernia

increases with time and the rate seem to increase for 8 to 10 years after

surgery [21, 34].

An incisional hernia is defined as abdominal contents protruding

through a defect in the abdominal wall. An ostomy is a bowel brought out

through a surgically created opening in the abdominal wall. Thus, it can be

argued that a surgeon creating a stoma per definition also creates an

incisional hernia. In view of this it is perhaps not surprising that a high

proportion of patients with an ostomy over time develop a parastomal hernia.

Incidence of Parastomal Hernia

The lack of a uniform definition of parastomal hernia used at follow-up makes

it difficult to estimate the true rate of herniation. It is very hard to make a true

estimate as in several reports the authors have not presented how

parastomal hernia was defined at follow-up examination [13-14, 17-18, 20-22,

26-28, 31, 34-53]. The incidence reported varies greatly and this is probably

very much related to different definition of parastomal hernia used at follow-

up examination. With a CT-scan the incidence of parastomal hernia has

sometimes been higher than at clinical examination. This has been attributed

to small parastomal hernias not noticed at clinical examination being detected

with a CT-scan [13-14, 22, 24]

15

Follow-up time varies considerably in clinical reports on parastomal

hernia and in only a few studies follow-up examination has been after a year

or more [17, 20-22, 26-27, 38-39, 50, 54]. The rate of parastomal hernia

reported varies from 5% up to 80% [13-14, 22, 24-28, 31, 34-37, 39-48].

Experiences gained in studies on incisional hernia indicate that the highest

hernia rates reported are the most accurate [11] . Considering that no uniform

definition of parastomal hernia has been used in available reports and the

variability of time to follow-up, the true rate of parastomal hernia is very

difficult to estimate but is it probably between 30% and 50% in general

surgical practice.

The rate of parastomal hernia has been suggested to be lower after

an ileostomy than after a colostomy [39, 44, 55-56]. There are reports that did

not detect any difference in herniation rates between ileostomies and

colostomies, so the matter is not definitely settled [13, 26, 57]. The rate of

herniation with loop ileostomies and loop colostomies cannot easily be

compared with end stomas. Loop stomas are often intended to be temporary

and bowel continuity is then often restored after a short period of time. Loop

stomas are also often used as a palliative means in patients with short

expected survival. Thus, time to follow-up is often very different between loop

stomas and end stomas [3-4, 32, 58-59].

When creating an ileostomy after a Bricker diversion the rate of

parastomal hernia has been reported within the same range as with other

ostomies i.e. 5% to 65% [54, 57, 60-64]

Surgical Technique When Forming a Stoma

Throughout the history of surgical stoma formation there have been a great

variety of methods used to bring out a stoma [1]. There are no randomized

trials available comparing different ways to bring out a stoma and there are

16

only retrospective reports at hand. For certain you should never bring out a

stoma through the laparotomy wound since it causes disastrous result in

terms of wound infection, wound dehiscence and herniation [2, 65-67]. An

extra-peritoneal route has in two retrospective studies been associated with a

lower rate of parastomal hernia than reported with a conventional straight

route but there is no randomized trial available and the definition of

parastomal hernia used at follow up was not presented [38, 68]. The extra-

peritoneal technique has on these grounds been advocated by some but it

has also been challenged by others and is today probably not widely used

[69-70].

In the Scandinavian countries it is probably quite common to bring out

the stoma through the rectus abdominis muscle. This practice is based on

one Swedish and one Danish retrospective non-randomized report of a lower

rate of parastomal hernia with the stoma brought out through the rectus

muscle than lateral to the rectus muscle [36, 44]. Four other retrospective

studies did not detect any difference in the rate of herniation with these

methods [13, 37-38, 71] and in one recent report a lower herniation rate was

found if the stoma was brought out lateral to the rectus muscle [72], The

definition of parastomal hernia at follow-up is not given in these clinical

reports and there is no randomized trial comparing the methods. The

indications of superior results with the stoma through the rectus muscle

available, that this places the stoma close to the midline - appreciated by

patients and stoma therapists at it facilitates stoma care - and that there is no

obvious disadvantage associated with this stoma placement, has probably

enhanced it acceptance among Scandinavian surgeons.

Making a too large opening in the abdominal wall for the stoma is

often claimed to be the cause of parastomal hernias developing [2, 27, 43,

67, 70, 73]. There is no clinical evidence available as of what is to be

17

regarded as a too large opening but a suggestion of not exceeding a

diameter of 2.5 cm has been made. In surgical practice it has to be

recognized that the opening in the abdominal wall must always be large

enough to let the bowel through and in case the bowel is very wide and

embedded in fat the opening must be sized accordingly. There is no evidence

that fixating the mesentery or suturing the bowel to the aponeurosis diminish

the rate of parastomal hernia [3, 26, 38-39]. There are some patient and

operative factors suggested to be associated with the development of

parastomal hernia. Thus, the risk for parastomal hernia has been proposed to

increase with old age, wound infection, obesity, corticosteroid use, chronic

respiratory disorders, malnutrition and increased waist circumference [2, 12,

71, 74-75].

With an ostomy created with a laparoscopic technique the parastomal

hernia rate has been reported to be between none and 12%. However,

follow-up was in these reports less than 12 months and there was no

definition of parastomal hernia given [76-77]. In a recent publication a

laparoscopic extra-peritoneal route was utilized when creating the stoma but

there was no follow-up of patients presented [78]. There are no randomized

trials comparing results with an open versus a laparoscopic formation of

stomas. In two clinical series with less than 25 patients included in each

report a trephine device was tried for the formation of stomas [79-80]

Surgical Treatment of Parastomal Hernia

Surgical repair has in retrospective reports been called for in 11% to 70% of

patients with a parastomal hernia [13, 21, 25, 27, 44, 54]. Once again the

lack of a uniform definition of parastomal hernia makes it difficult to deal with

this figure. Available reports within this field have not presented the definition

of parastomal hernia used. With a wide definition of parastomal hernia and a

18

consequent high rate of herniation many hernias may be small and a small

proportion might demand repair. With a narrow definition fewer but larger

hernias may be detected and a higher proportion might require repair. There

are no randomized studies comparing results with different surgical

treatments of parastomal hernias and only small series with a particular

technique from single centres are available [4] .

There are four strategies when repairing parastomal hernias: local

repair, relocation of the stoma, mesh repair and relocation of the stoma with a

prophylactic mesh at the new stoma site. Local aponeurotic repair is today an

obsolete method since it has produced an unacceptably high recurrence rate,

reported in the range of 10% to 76% [2-3, 12, 81-85]. After relocation of the

stoma the risk of developing a new parastomal hernia is at least as high as

after the first operation and recurrence rate of 24% to 86% have been

reported [12, 81-82, 84, 86-87]. If the stoma is relocated a second time

recurrence rates are reported to be even higher [81]. If a stoma is relocated it

should be relocated to the other side of the midline since recurrence rates

have been reported to be very high if relocation is to the same side of the

midline [84].

The lowest recurrence rates have been reported with prosthetic mesh

repairs, although there is no randomized trial available comparing mesh

repair with local repair or relocation of the stoma. The strategy of repairing

parastomal hernias with a mesh is based on the experiences with mesh

repair of incisional hernias [88]. Mesh repair is a common and well

established

technique for the repair of incisional hernias [89-90]. The mesh can be placed

in an onlay, an inlay, a sublay or an intra peritoneal onlay position (IPOM)

[73-74, 88, 91]. The IPOM technique can be used in both open and

laparoscopic repairs.

19

There are several types of meshes available: non absorbable,

absorbable [88], partly absorbable [88] and acellular collagen matrix [92]

meshes, all of which have been used for parastomal hernia repair. Several

types of meshes can be placed in a contaminated environment without major

complications [17, 93]. There are though potential dangers associated with

the use of meshes, such as fistula formation, adhesions, septic complications

and seroma formation [29, 94-95]. Meshes that induce an inflammatory tissue

response cannot be placed in contact with abdominal contents without a risk

of fistula formation, adhesions and septic complications [96]. With the IPOM

technique a mesh constructed of two layers is therefore usually used. The

mesh surface facing the abdominal wall is usually of a non absorbable

material inducing tissue response allowing integration of the mesh to the

abdominal wall. The mesh surface facing abdominal contents is a non

reactive material causing a low or negligible inflammatory response so that

adhesions are not formed.

Relocation of the stoma to the other side of the midline with a

prophylactic mesh at the new stoma site and a mesh repair at the old stoma

site has been reported in one small series with few complications and no

recurrence detected after 12 months [97].

Prevention of Parastomal Hernia

Since parastomal hernia is a common complication that is difficult to treat,

several prophylactic measures have been tried. Placing a prosthetic mesh at

the time of the initial stoma formation was first suggested by Rosin and

Bornardi in 1977 [98]. In view of the similarities between incisional hernias

and ostomies, using a prophylactic mesh may appear rational when creating

a stoma. Incisional hernia is defined as intra abdominal contents protruding

20

through a defect in the abdominal wall. Constructing an ostomy the surgeon

creates a defect in the abdominal wall for the bowel to pass through. As this

per definition is a hernia the high rates of parastomal hernia encountered are

perhaps not surprising. If ostomies are regarded as intentionally created

hernias it seems logical that parastomal hernias be prevented in the same

way as incisional hernias are repaired i.e. with a mesh. Bayer published the

first clinical series with a prophylactic prosthetic mesh in 1986 [99]. In that

study, 36 patients received a polypropylene ring with four cross wise

prolongations fashioned as a fascia onlay and no herniation was reported

after 4 years. Local infection developed in three patients but this did not

prompt removal of the mesh. After 8 months one patient developed a stricture

and the mesh had to be removed. Light described a technique for intra

peritoneal mesh placement at the time of the colostomy formation but

provided no data on the rate of complications [100].

New prosthetic meshes have been developed with a large pore size of

about 5 millimetres. Such low weight, large pore meshes are associated with

a low degree of inflammatory response in the vicinity of the mesh [101]. This

probably reduces the risk for the mesh eroding into the bowel [88]. Low

weight, large pore meshes that are constructed out of equal proportions of

non-absorbable and absorbable materials are also available. The rational for

this construction is that the mesh is initially heavy enough to induce tissue

response for adequate tissue integration but is still strong enough to achieve

its purpose when the absorbable part of the mesh has vanished.

21

Types of Meshes

A high weight, small pore mesh is dense and induces a considerable tissue

response.

A low weight, small pore mesh is not as dense as a high weight mesh but the

pore size is small. The mesh induces less tissue response than a high weight

mesh.

A low weight, large pore mesh is as dense as a light weight mesh but is

constructed with large pores. The tissue response is similar to a low weight

mesh but there is minimal tissue response within the open pores.

22

Aims of the Studies

The aims of the studies are

1. to evaluate the short and long term effects of a prophylactic

prosthetic mesh placed in a sublay position at the index

operation on the development of parastomal hernia

2. to evaluate the short and long term effects a prophylactic mesh

on surgical site infection, stoma stenosis and enteric fistula

formation

3. to validate a definition of parastomal hernia at clinical

examination

4. to validate a method and a definition of parastomal hernia at CT-

scan examination

23

Patients and Methods

Articles I and II

Between January 2001 and April 2003 patients scheduled for an ostomy

were randomised to either a conventional stoma through the rectus

abdominis muscle or to the same procedure with the addition of a prosthetic

mesh. Randomisation was by opening consecutively numbered sealed

envelopes, arranged into batches of four. The mesh was placed in a sublay

position between the rectus abdominis muscle and the posterior rectus

sheath (Figure 1). A Vypro® (Ethicon, Norderstedt, Germany) mesh was cut

to 10 by 10 centimetres, and the bowel was brought out through a cross 2 by

2 centimetres cut in its centre. The mesh was anchored to the posterior

rectus sheath with absorbable stitches placed in the lateral corners. Closing

the midline incision the running suture included the medial part of the mesh

together with peritoneum to prevent the mesh coming into contact with

abdominal contents. Patient age, sex and body mass index (BMI = weight /

length2) were recorded. The indication for surgery and if surgery was

emergent were noted. In elective surgery prophylactic antibiotics were a

single oral dose of tetracyklin and metronidazol. In emergent surgery

intravenous antibiotics were administered dependant on the clinical situation.

Patients were examined after 1 month, 12 months and 5 years.

Patients were examined relaxed and straining in an erect position and

relaxed and with lifted legs in a supine position. Any protrusion in the vicinity

of the stoma was considered a parastomal hernia. Wound infection, mesh

infection, fistula formation, stricture formation, reoperation and parastomal

hernia repair were recorded. An independent observer (Inga-Lena Nilsson,

MD, PhD) had the power to terminate the trial if the annually analysed

complication rates proved unacceptably high.

24

The study was approved by the regional ethics committee. The

SPSSTM

18.0 (SPSS Inc, Chicago, Illinois, USA) software was used for

statistical analysis. Fisher´s exact test and Mann-Whitney U test were used

for the analysis as appropriate. ClinicalTrials.gov ID: NCT00509054.

Article III

Between April 2003 and November 2006 a prophylactic mesh was intended

in all patients with an end ostomy created at open surgery. The abdomen

was accessed through the midline. A partially absorbable low-weight mesh

was placed in a sublay position between the rectus abdominis muscle and

the posterior rectus sheath (Figure 1). The bowel was brought out through a

cross cut in the centre of a 10 by 10 centimetres Ultrapro ® mesh (Ethicon,

Norderstedt, Germany). The cross was cut just large enough to let the bowel

through. The mesh was anchored to the posterior rectus sheath with

absorbable stitches placed in its lateral corners. The running suture closing

the midline included the medial part of the mesh together with peritoneum to

prevent the mesh coming in contact with abdominal contents. In elective

surgery prophylactic antibiotics were a single oral dose of tetracyklin and

metronidazol. In emergent surgery intravenous antibiotics were administered

dependant on the clinical situation.

Patient age, sex and body mass index (BMI = weight / length2) were

recorded. The indication for surgery and if surgery was emergent were noted.

Wound contamination was registered and wounds were regarded dirty if

there was pus or faecal contents in the abdominal cavity. Surgical site

infection was recorded ten days after surgery, defined according to the

definition of the US Centre for Disease Control. One year after surgery

patients were examined by one surgeon (AJ). Surgical site infection,

25

parastomal hernia, fistula formation, stenosis of the stoma and if the mesh

had been removed were recorded. A protrusion in the vicinity of the stoma

with the patient straining in a supine and an erect position was regarded as a

parastomal hernia.

The SPSSTM

18.0 (SPSS Inc, Chicago, Illinois, USA) software was

used for the statistical analysis. Mean and 95% confidence intervals (CI) were

calculated.

Article IV

Patients with a sigmoidostomy included in the routine follow-up program for

colorectal cancer used at the department were first assessed by one of the

surgeons (AJ) for presence of parastomal hernia. Inclusion in the study aimed

at achieving an even distribution between parastomal hernias and intact

stomas. Patients were then clinically examined by another senior and one

young surgeon. All three raters were at clinical examination unaware of any

previous judgment of the presence of parastomal hernia or of the use of a

prophylactic mesh. The definition of parastomal hernia used at clinical

examination was thoroughly discussed among the raters prior to patient

examination.

The definition of parastomal hernia at CT-scan was discussed and

decided upon by the three authors (AJ, LW, LAI) prior to the assessment of

CT-scans. The senior radiologist (LW) thoroughly informed and discussed the

definition to be used with the other two radiological raters prior to their

ratings. All radiological raters were at examination unaware of any previous

clinical judgment of the presence of parastomal hernia. All three radiologists

examined CT-scans independently of each other.

26

The maximum distance between aponeurotic edges at the stoma site

was measured in the vertical and the horizontal planes. The area of the

stoma opening was calculated as measurements being diagonals of a rhomb

(height x width / 2).

Parastomal hernia was at clinical examination defined as any

protrusion in the vicinity of the stoma with the patient straining in a supine

and an erect position. At CT-scan parastomal hernia was defined as any intra

abdominal content protruding beyond the peritoneum or the presence of a

hernia sac. The conventional CT-scan was performed with the patients

supine. With the new CT-scan method patients were in the prone position

with the stoma positioned in the centre of an inflatable plastic ring

approximately 10 centimeters high. Adding the novel CT-scan examining

patients in prone position was approved by the local Radiation committee.

This examination was voluntary to the patients and they all agreed.

The SPSSTM

18.0 (SPSS Inc, Chicago, Illinois, USA) was used for the

analysis. Fleiss Kappa for multiple raters was used utilizing a macro

developed for SPSS [102-103].

27

Figure 1. A prophylactic prosthetic mesh is placed in a sublay position,

posterior to the rectus abdominis muscle and anterior to the dorsal rectus

sheath.

28

Results

Article I and II

Fifty-four patients were included in the randomised study. 27 were allocated

to receive a conventional stoma, and 27 to the addition of a prophylactic

mesh (Table 1). In the group with a conventional stoma one patient died

before 12-month follow-up and parastomal hernia was present in 13 of the 26

remaining. In the group with a prophylactic mesh six patients died before 12-

month follow-up and parastomal hernia was present in 1 of the 21 remaining

(Table 2). With these results available the independent observer stopped the

randomized trial as it was judged to be unethical to continue the study when it

became evident that use of the mesh involved a dramatically lower rate of

parastomal hernia. It was the observer’s view that it was unethical to subject

patients not receiving a mesh to the very high rate of parastomal hernia.

Five year follow up was after 57 to 83 months (mean 65.2), similar for

both groups. After 5 years another 5 patients with a conventional stoma had

died. In the 21 remaining parastomal hernia was found in 12 and 5 had been

subjected to parastomal hernia repair. In the group receiving a prophylactic

mesh another six patients had died and in the 15 remaining parastomal

hernia was present in 2, which had not demanded hernia repair (p<0.001).

(Table 2)

No fistula formation or stoma stricture was found in the two groups

after one and five years. No mesh infection was noted and no mesh was

removed during the study period.

29

Table 1. Patient and operative characteristics related to enterostoma with

or without a prosthetic mesh

No mesh

n=27

Mesh

n=27

p

value

Age, mean years (95% CI)

71 (67-76) 70 (64-75) 0.80 §

Women, No.

11 12 1.0 *

Body Mass Index, mean (95% CI)

27 (25-29) 26 (24-28) 0.56 §

Emergency laparotomy

4 1 0.35 *

Reason for operation:

Malignant tumour

22 25 0.42 *

Diverticulitis

3 1 0.61 *

Inflammatory bowel disease

1 1 1.0 *

Other

1 0 1.0 *

* Fisher’s exact test, § Mann Whitney U-test

30

Table 2. Parastomal hernia at 1- and 5-year follow-up

No mesh

n=27

Mesh

n=27

p

value

Died before 1-year follow up

1

6

0.10 *

Parastomal hernia after 1 year

13 of 26 1 of 21 <0.01*

Died between 1-year and 5-year

follow up

5 6 0.51*

Mean follow up (months)

64.5 66.0 0.80 §

Parastomal hernia after 5 years 17 of 21 2 of 15 <0.01 *

Fistula, n

0 0

Stenosis, n

0 0

Mesh removed, n

0 0

* Fisher’s exact test, § Mann Whitney U-test

31

Article III

An end ostomy was constructed in 93 patients. The most common indication

for surgery was malignant disease. In 8 patients other indications were for

example anal incontinence, rectal fistula, perineal trauma and obstetric

complications. (Table 3 - 4)

A prophylactic mesh was used in 75 patients and was omitted in 18. A

mesh was omitted in 8 of 68 patients (12%) operated on by four surgeons

that performed at least 10 operation during the study period. A mesh was

omitted in 10 of 25 patients (40%) by 8 surgeons performing less than 10

operations. In 9 patients a mesh could not be placed at the planned stoma

site due to severe scarring of peritoneum and the abdominal wall after

previous surgery. The surgeon omitted a prophylactic mesh in 5 patients

because of limited life expectancy and in 4 patients because of planned early

stoma closure.

A prophylactic mesh was used in 19 of 28 (68%) emergent operations

and in 56 of 65 (86%) elective. A prophylactic mesh was used in 19 of 29

(65%) dirty wounds and in 56 of 64 (87%) contaminated. In patients receiving

a mesh mean age was 65 (95% CI 62, 68) years and 57 (95% CI 47, 68)

without a mesh.

Within 10 days of surgery 5 patients died of septic or cardiovascular

complications not related to the use of a mesh. Surgical site infection was

present in 10 of the 88 (11%) remaining. (Table 4) Infections were in the

laparotomy wound and no mesh was infected. Infection was more common

when a mesh was omitted. The rate of the infection was similar for surgeons

performing less than 10 operations and those creating more stomas (10% vs.

12%). Most infections were minor, not requiring treatment. In two patients a

32

major

infection

demanded

antibiotic

treatment

and surgical

drainage.

There were

no late

infections or

persistent

infections

during the

observation

period.

Table 3. Patient and operative characteristics for 93 patients

with an ostomy created

N 93

Age, mean years (95% CI)

63 (60, 67)

Male / Female

61 / 32

Body Mass Index, mean (95% CI) 25 (24, 26)

Malignant disease 65

Inflammatory bowel disease 7

Perforated diverticulitis 10

Other 8

Emergent surgery 28

Degree of wound contamination:

Contaminated 64

Dirty 29

Ostomy:

Colon 79

Ileum 14

33

Table 4. Surgical site infection in 88 patients operated on with an ostomy

created

No mesh

Mesh

Surgical site infection, n (95% CI) 4 of 15 (27% 1, 52) 6 of 73 (8% 2, 15)

Minor infection, n

3

5

Major infection, n

1 1

Wound contaminated

Surgical site infection, n

1 of 7 3 of 55

Wound dirty

Surgical site infection, n

3 of 8 3 of 18

34

Before one year follow-up 13 patients died and in 7 bowel continuity had

been restored. Follow up examination was for the 73 remaining after 10 to 45

months (mean 15 months, 95% CI 14, 16). With a prophylactic mesh

parastomal hernia was less frequent than without a mesh. (Table 5)

Table 5. Complications after one year in 73 patients operated on with an ostomy

created

No mesh

Mesh

Parastomal hernia, n (95% CI) 8 of 12 (67% 35, 98) 8 of 61 (13% 4, 22)

Colostomy

7 of 8

7 of 52

Ileostomy

1 of 4 1 of 9

Fistula, n

0 0

Stenosis, n

0 0

Mesh removed, n

0 0

35

Article IV

In 27 patients the mean area of the stoma opening was 4.3 cm2 (Range 0.6

cm2 to 11.3 cm

2). At CT-scan in the prone position assessments were in total

agreement for surgeons and radiologists in 21 patients. They then regarded

11 parastomal hernias to be present and 10 stomas to be intact. In these 21

patients herniation was regarded present in none of 8 patients with a stoma

area of less than 3 cm2 and in 11 of 13 with a larger opening (p<0.001).

In 6 patients the raters’ assessment was not in total agreement (Table

6). In patient A surgeons regarded the stoma to be intact and radiologists

considered a parastomal hernia to be present. In patients B and C (stoma

area of 0.9 cm2 and 1.9 cm

2 respectively) only one radiologist regarded a

parastomal hernia to be present.

In 4 patients a parastomal hernia was detected at CT-scan with the

patient in the prone position but not when supine. Stoma area was in these

patients 3.1 cm2, 3.5 cm

2, 4.3 cm

2 and 8.3 cm

2 respectively. In 3 out of these

there was total agreement on a parastomal hernia being present among the

surgeons and radiologists with CT-scan in the prone position (Table 6). In

one patient two surgeons did not detect a parastomal hernia.

For the surgeons at clinical examination Kappa was 0.85. For the

radiologists Kappa was 0.85 with CT-scan in the prone position and 0.82 in

the supine position. For the surgeons and radiologists collectively, Kappa was

0.80 with CT-scan in the prone position. With CT-scan in the supine position

Kappa was for the surgeons and radiologists collectively 0.64. With CT-scan

in the prone and supine position Kappa was 0.65. The assessments of the

two senior surgeons were in total agreement (Kappa 1.0) and together with

the two radiologists that differed least between each another (Kappa 0.93)

the collective Kappa was 0.90 was for CT-scan in prone position. (p<0.001

for all Kappa values)

36

Table 6. Raters’ assessment of a parastomal hernia being present in 6

patients where there was not total agreement for all surgeons and

radiologists, at CT-scan in prone position.

Surgeon

Radiologist

Patient

1

2

3

1

2

3

A

No

No

No

Yes

Yes

Yes

B No No No No No Yes

C No No No No No Yes

D No No Yes No No No

E No No Yes Yes Yes Yes

F Yes Yes No No Yes No

37

38

Discussion

Parastomal hernia is a clinical problem that despite several efforts to find

surgical techniques to lower the rates continues to be a major problem. An

incisional hernia is defined as abdominal contents protruding through a

defect in the abdominal wall. This means that the surgeon constructing an

ostomy per definition also creates an incisional hernia. Experiences during

the last decades clearly show that in order to achieve low recurrence rates

incisional hernias should be repaired with a prosthetic mesh. If one regards

an ostomy as a deliberately created incisional hernia the very high rates of

parastomal hernia are perhaps not surprising. Continuing this line of thought

it may also seem logical that the path towards preventing the development of

parastomal hernia is similar to the way incisional hernias are repaired i.e. with

a prosthetic mesh.

There are several reasons to hesitate in placing a prosthetic mesh at

the stoma site. The clinical experience is that a mesh in close contact with

bowel may cause adhesions, inflammatory response, strictures, fistulas and

severe septic complications. These complications are to some extent related

to the properties of the mesh. High density, small pore meshes induce a

higher degree of inflammatory response in the tissues and are prone to erode

into the bowel and cause severe septic complications. In general hernia

surgery these aspects have been considered for many years and there is a

distinct trend towards utilizing low weight, large pore meshes. Such meshes

induce a lower degree of inflammatory response and therefore probably a

lower risk of strictures, adhesions and erosion into the bowel. Integration of

the mesh into surrounding tissues seems to be good enough for the stability

of the repair, with available low weight meshes. The concept of low weight,

large pore meshes is presently not quite clear as there is no uniform

39

definition or standardisation of mesh properties. There is on the other hand

considerable clinical and experimental experience available for a great variety

of meshes.

The Vypro® (Ethicon, Norderstedt, Germany) mesh is generally

regarded as a low weight, large pore mesh. It is constituted out of equal

proportions of monofilament polypropylene and multifilament polyglactin.

Succeeding the Vypro® mesh the Ultrapro® (Ethicon, Norderstedt, Germany)

mesh was developed, constituted out of equal proportions of monofilament

polypropylene and monofilament poliglecaprone. The polypropylene

component is non absorbable and after integration of the mesh into

surrounding tissues constitutes the long term stability of the mesh. The

polyglactin and poliglecaprone components are absorbable materials

contributing to the strength of the mesh for two to three weeks and are within

a couple of months completely absorbed. The rational for this construction is

that the mesh initially has a higher density inducing sufficient inflammatory

response for integration of the mesh into surrounding tissues and as this

phase is completed only the proportion of the mesh necessary for its

supportive function is left in the wound. Generally materials implanted in

tissues should be monofilament as they are associated with a lower risk of

infection than multifilaments. For this reason the multifilament polyglactin in

the Vypro® mesh was in the Ultrapro® mesh replaced by the monofilament

poliglecaprone. [104-105]

There is considerable clinical experience with the use of low weight,

large pore meshes. It seems to be possible to place such meshes into close

contact with bowel without detrimental effects and their supportive effect has

been sufficient for inguinal and incisional hernia repair. They appear resistant

to infection and if an infection occurs the mesh does not have to be removed

as infection subsides with general surgical and antibiotic treatment of surgical

40

site infection. Searching for a suitable prosthetic mesh to place into close

contact with bowel low weight, large pore meshes seemed suitable and were

therefore chosen for the present studies.

Although theoretical assumptions and clinical experiences were in

favour of placing a low weight, large pore mesh as a prophylactic mesh in

close contact with bowel at stoma formation a degree of uncertainty

regarding complications still remained. There was a possibility of stricture,

mesh eroding into bowel and severe septic complications developing within

this setting. To avoid that in our randomized study the trial was continued

despite such complications frequently occurring an independent observer

annually looked into the number of complications recorded and had the

power to close randomization if an unduly high rate occurred. The observer

did eventually close randomization. Not because of a high rate of

complications with a mesh utilized but because it was considered unethical to

subject patients to a very high rate of parastomal hernia when a mesh was

not used.

The intension was to include all patients at the department provided

with an ostomy into the randomized trial. Surgeons and staff were informed

about the trial, inclusion criteria and of the surgical technique when placing a

prophylactic prosthetic mesh. The trial was approved by the local ethics

committee. Patients were informed and were asked for their consent for

inclusion in the trial. Generally all patients asked approved inclusion but in the

emergent situation many patients could not be included as the situation and

their septic condition was not compatible with obtaining proper informed

consent. Randomization was by opening closed envelopes. These were

sorted in batches of four as it was regarded important for surgeons’

adherence to the trial that a mesh was utilized on a regular basis. This had

the consequence that when multiples of four were randomized an equal

41

number of mesh / non mesh was included. When an even number, not a

multiple of four, were randomized the probability of an equal number of mesh

/ non mesh was 66%.

Results of the randomized trial after one year and after five years

convincingly showed that with a low weight, large pore mesh placed in a

sublay position at the index operation substantially reduces the risk of

parastomal hernia. This was achieved without any long term complications

related to the mesh [16, 21]. These findings are in congruence with two later

randomized trials [22, 52] and several clinical reports [19-20, 50, 52-53, 106]

with a prophylactic mesh utilized in sublay, onlay or IPOM position. One of

the randomized trials was conducted, into detail, similarly to the present trial

and results were also very comparable. It is very interesting that a mesh in

the present setting could be placed in severely contaminated surroundings

with an acceptable rate of wound infection and without infection of the mesh

occurring [107]. The sublay position of the mesh may have been important for

this outcome as it was then into close contact with a well vascularised rectus

abdominis muscle possible enhancing rapid tissue integration of the mesh

and prompt cellular response.

After closing the randomized trial a prophylactic mesh was used in all

end stomas at the department. A prophylactic mesh utilized in routine surgery

with several surgeons involved and in both elective and emergency surgery

was evaluated. This study currently represents the largest cohort using a

prophylactic mesh when forming a stoma. Findings in this study showed that

a prophylactic mesh could be placed in most patients. A mesh had to be

omitted in only 10% because of severe scarring of the abdominal wall after

previous surgery. It was also obvious that surgeons at the index operation

were not able to anticipate the patients´ expected survival or the probability of

a permanent stoma. Thus, surgeons tended to omit a prophylactic mesh in

42

patients that may very well have profited by a prophylactic mesh. This

tendency was more pronounced for surgeons inexperienced in colorectal

surgery. The rate of parastomal hernia was in routine surgery very similar to

those achieved in the randomized trial. It is very interesting that a mesh could

also be placed in a large number of patients with severely contaminated

wounds, including faecal peritonitis, with an acceptable rate of infection and

without infection of the mesh occurring.

The question must be raised of why a prophylactic prosthetic mesh

reduces the rate of parastomal hernia. The clue to answering this question

may be to recognize that there was a strong correlation between the area of

the stoma opening and the presence of herniation at CT-scan. The opening

made by cutting a cross in the centre of the mesh probably sets an upper

limit for the size of the stoma opening. Thus, a prophylactic mesh may

prevent the development of parastomal hernia because it prevents the stoma

opening from growing any larger than the size created at the index operation.

The lack of a uniform definition of parastomal hernia at clinical

examination and at CT-scan is a major problem when comparing results in

different reports. In the five year follow-up of the randomized trial 80 % of

patients had developed a parastomal hernia when a conventional stoma

without a prophylactic mesh was constructed. That almost every patient with

a stoma actually develops a parastomal hernia is difficult to accept and may

have affected the definition of herniation used at follow-up. In reports with a

low rate of parastomal hernia the authors may, reluctant to accept the very

high rates otherwise encountered, have regarded small protrusions as a

normal condition after stoma formation and regarded only larger protrusions

as parastomal hernias. Concerning this no exact knowledge is possible

regarding studies reported before 2004 since the definition of herniation used

at follow-up was not presented. Beginning in 2004 many authors have used

43

the same clinical definition as in the present studies and they have reported

very similar herniation rates.

In our studies a very standardized definition of parastomal hernia was

used at clinical examination. Any protrusion in the vicinity of the stoma with

the patient straining in both supine and erect positions was regarded as a

parastomal hernia. With this simple definition there is a minimum of

subjectivity when grading of the protrusion. In order to validate this definition

three surgeons independently examined patients and achieved a Kappa of

0.85. Kappa 0.60 to 0.79 is generally regarded as a good correlation between

raters´ judgment and Kappa 0.80 or higher as a very good correlation. CT-

scan is usually done with the patient in the supine position. It is then difficult

to ask the patients to strain during the time it takes to perform the CT-scan

and it cannot be performed with the patient erect. A new method was

therefore developed adding a CT-scan with the patient in the prone position

and the stoma placed in the centre of an inflatable ring. In this way a situation

with the patient constantly straining was mimicked as all intra abdominal

pressure was transmitted to the stoma. Thee radiologists independently

examined the CT-scans. For radiologists Kappa was 0.85 with CT-scan in the

prone position and 0.82 in the supine position. For surgeons and radiologists

collectively, Kappa was 0.80 with CT-scan in the prone position and 0.64 in

the supine position. When all surgeons agreed that there was a parastomal

hernia present at clinical examination all radiologists also detected a hernia at

CT-scan. Thus, there was a strong correlation and reproducibility for the

clinical and the radiological definitions used. This was very probably related

to the simplicity of the definitions and the thorough information and

discussion of the definitions used among raters before the assessments were

done. The correlation between clinical examination and findings at CT-scan

with the patient’s in the prone position was very strong. It was not as strong

44

with CT-scan in the supine position and it was obvious that several

parastomal hernias that were detected at clinical examination and at CT-scan

in the prone position could not be detected with patients supine

The strong correlation between clinical examination and CT-scan in

the prone position support the theory that the clinical definition used at follow-

up examination in the present studies is relevant. Kald has previously shown

that patients with a parastomal hernia according to this definition have a

poorer quality of life [5]. In the present study we found a very strong

agreement between the clinical diagnosis and the radiological diagnosis. It

was in fact so strong that a CT-scan as a complement to the clinical

examination hardly seems justified. Thus, the present and other reports that

have used this clinical definition have not only reported very similar rates of

parastomal hernia but have probably also reported a rate very much

reflecting the clinical reality.

It has been claimed that with a CT-scan small parastomal hernias

are detected that cannot be found clinically [13-14, 22]. This was not the case

in the present study when patients were examined in the prone position. It

was also obvious that several hernias detected at clinical examination could

not be detected with a CT-scan with patients supine. There have been

attempts to subtype parastomal hernias according to how they present at CT-

scan but in the present context this did not serve any purpose. The aim of the

present studies was to find and validate clinical and radiological definitions of

parastomal hernia that distinguished between the presence and absence of a

parastomal hernia [3-4, 97, 108-109].

In the present studies a validated definition of parastomal hernia at

clinical examination and with CT-scan was found. A prophylactic prosthetic

mesh reduced the rate of parastomal hernias developing without increasing

the rate of complications. It seems appropriate to conclude that the present

45

studies show that the path towards reducing the rate of parastomal hernia

includes the use of a prophylactic mesh.

Future perspectives

We cannot know, however, if the type of mesh used in the present studies is

the optimal mesh or if other types of meshes can be used for the same

purpose with similar results. Future studies may identify meshes or new

meshes may be developed that are even more suitable in this situation.

Meshes were placed in a sublay position. This placement was quick and easy

to perform and it can also be done by a laparoscopic technique with similar

results [110].

Ileostomies are also constructed in urological procedures such as

Bricker diversions, probably with similar high parastomal hernia rates as with

other ostomies. A major problem with parastomal hernias after Bricker

diversion is that the repair of such hernias is technically very difficult and

represents a major surgical challenge. The recurrence rate after such repair

is not known. At the Department of Urology, Sundsvall hospital, a

prophylactic low weight, large pore mesh in a sublay position has also been

utilized for preventing the development of parastomal hernia after Bricker

diversion. Non published results of 36 ileostomies at Bricker diversions

receiving a prophylactic mesh indicate that similar low parastomal hernia

rates are achieved as with other ostomies.

Future studies may find that the mesh can be placed in other positions

with similar results. The most attractive positioning is then probably an intra

abdominal onlay position (IPOM) since it is technically an easy procedure.

This demands meshes, however, that provide appropriate tissue integration

of the mesh without increasing the risk of adhesions or the development of

46

enteric fistulas. Unfortunately long term effects of intra peritoneal meshes are

today virtually unknown. There are several meshes available with the mesh

surface facing abdominal contents constructed in a fashion that allegedly

prevents adhesions and inflammatory response. The problem is that long

term follow-up or randomized trials are lacking.

In the present studies a flat mesh with a cross cut in its centre was

used. It is very tempting to assume that meshes shaped with a funnel arising,

as developed by Berger [53] would further reduce the rate of parastomal

hernias developing. This is of course because a parastomal hernia, once the

mesh is integrated with surrounding tissues, can only protrude through the

space between the bowel and the mesh opening. If the flat portion of the

mesh continues into the funnel this weak spot for hernia development might

be eliminated. There are, however, several challenges encountered

constructing such a mesh. The shape of the funnel is especially crucial, as is

the type of mesh used for the funnel. If the funnel part of the mesh causes

too much tissue response adhesions, enteric fistulas and stenosis may prove

to be a problem. No doubt in the future there will be experiments and clinical

studies carried out in this direction.

The use of a prophylactic mesh in Bricker diversions should be

further explored and long term results be provided. Permanent stomas are

also constructed in children. With an ostomy present for their entire life it is of

course essential that no parastomal hernia develops. Although there is no

data available on the rate of parastomal hernia in children there is probably

no reason to assume the rate to be very much lower than in adults. Thus,

children may also benefit by the use of a prophylactic mesh. Generally, one

should be cautios about placing a prosthetic mesh in growing persons. This

area would therefore present a challenge for future studies.

47

Conclusions

1. A prophylactic prosthetic large pore, low weight mesh placed in a

sublay position at the index operation substantially reduces the rate of

parastomal hernia.

2. A prophylactic mesh is not associated with an increased rate of

surgical site infection, stoma stenosis or enteric fistulas, even if placed

in severely contaminated wounds.

3. Parastomal hernia should at clinical examination be defined as any

protrusion in the vicinity of the stoma with the patient straining in a

supine and an erect position. With this definition there is a strong

correlation between the assessments of clinical raters and CT-scan.

4. At CT-scan parastomal hernia should be defined as any intra

abdominal content protruding beyond peritoneum or the presence of a

hernia sac with the patient examined supine face down. With this

definition there is a strong correlation between the assessments of

radiological raters and clinical assessments.

48

References

1. Pearl, R.K., Parastomal hernias. World J Surg, 1989. 13(5): p. 569-72.

2. Goligher, J.C., Surgery of the Anus, Colon and Rectum (5th edn.).

1984, Baillière Tindall: London. p. 703-704.

3. Carne, P.W., G.M. Robertson, and F.A. Frizelle, Parastomal hernia. Br

J Surg, 2003. 90(7): p. 784-93.

4. Israelsson, L.A., Parastomal hernias. Surg Clin North Am, 2008. 88(1):

p. 113-25, ix.

5. Kald, A., et al., Quality of life is impaired in patients with peristomal

bulging of a sigmoid colostomy. Scand J Gastroenterol, 2008. 43(5):

p. 627-33.

6. Prieto, L., H. Thorsen, and K. Juul, Development and validation of a

quality of life questionnaire for patients with colostomy or ileostomy.

Health Qual Life Outcomes, 2005. 3: p. 62.

7. Pringle, W. and E. Swan, Continuing care after discharge from

hospital for stoma patients. Br J Nurs, 2001. 10(19): p. 1275-88.

8. Black, P., Managing physical postoperative stoma complications. Br J

Nurs, 2009. 18(17): p. S4-10.

9. Norton, C., J. Burch, and M.A. Kamm, Patients' views of a colostomy

for fecal incontinence. Dis Colon Rectum, 2005. 48(5): p. 1062-9.

10. Nilsson, E. and P. Nordin, Kirurgi. 2005. 2:a Upplagan: p. 211-224.

11. Cengiz, Y. and L.A. Israelsson, Incisional hernias in midline incisions:

an eight year-follow up. Hernia, 1998. 2: p. 175-177.

12. Devlin, H.B. and A.N. Kingsnorth, Parastomal hernia, in Management

of abdominal hernias, H.B. Devlin and A. Kingsnoth, Editors. 1998,

Butterworths: London. p. 257-266.

49

13. Williams, J.G., et al., Paraileostomy hernia: a clinical and radiological

study. Br J Surg, 1990. 77(12): p. 1355-1357.

14. Cingi, A., et al., Enterostomy site hernias: a clinical and computerized

tomographic evaluation. Dis Colon Rectum, 2006. 49(10): p. 1559-63.

15. Cingi, A., et al., Enterostomy closure site hernias: a clinical and

ultrasonographic evaluation. Hernia, 2008. 12(4): p. 401-5.

16. Janes, A., Y. Cengiz, and L.A. Israelsson, Preventing parastomal

hernia with a prosthetic mesh. Arch Surg, 2004. 139(12): p. 1356-8.

17. Janes, A., Y. Cengiz, and L.A. Israelsson, Randomized clinical trial of

the use of a prosthetic mesh to prevent parastomal hernia. Br J Surg,

2004. 91(3): p. 280-2.

18. Lopez-Cano, M., R. Lozoya-Trujillo, and E. Espin-Basany, Prosthetic

mesh in parastomal hernia prevention. Laparoscopic approach. Dis

Colon Rectum, 2009. 52(5): p. 1006-7.

19. Marimuthu, K.V., C. Ghosh, D. Mathew, Prevention of parastomal

hernia using preperitoneal mesh: a prospective observational study.

Colorectal Dis, 2006. 8(8): p. 672-675.

20. Vijayasekar, C., et al., Parastomal hernia: is prevention better than

cure? Use of preperitoneal polypropylene mesh at the time of stoma

formation. Tech Coloproctol, 2008. 12(4): p. 309-13.

21. Janes, A., Y. Cengiz, and L.A. Israelsson, Preventing parastomal

hernia with a prosthetic mesh: a 5-year follow-up of a randomized

study. World J Surg, 2009. 33(1): p. 118-21; discussion 122-3.

22. Serra-Aracil, X., et al., Randomized, controlled, prospective trial of the

use of a mesh to prevent parastomal hernia. Ann Surg, 2009. 249(4):

p. 583-7.

23. Gurmu, A., Parastomalt bråck, en klinisk eller radiologisk diagnos?

2009: p. 26 - F 81 Kolorektal - Kirugveckan 2009.

50

24. Moreno-Matias, J., et al., The prevalence of parastomal hernia after

formation of an end colostomy. A new clinico-radiological

classification. Colorectal Dis, 2009. 11(2): p. 173-7.

25. Burns, F.J., Complications of colostomy. Dis Colon Rectum, 1970.

13(6): p. 448-450.

26. Makela, J.T., P.H. Turko, and S.T. Laitenen, Analysis of late stomal

complications following ostomy surgery. Ann Chir Gynaecol, 1997.

86(4): p. 305-310.

27. Kronborg, O., et al., Late complications following operations for

cancer of the rectum and anus. Dis Colon Rectum, 1974. 17: p. 750-

753.

28. Arumugam, P.J., et al., A prospective audit of stomas--analysis of risk

factors and complications and their management. Colorectal Dis,

2003. 5(1): p. 49-52.

29. Steele, S.R., et al., Is parastomal hernia repair with polypropylene

mesh safe? Am J Surg, 2003. 185(5): p. 436-40.

30. Hasegawa, H., K. Yoshioka, and M.R. Keighley, Randomized trial of

fecal diversion for sphincter repair. Dis Colon Rectum, 2000. 43(7): p.

961-4; discussion 964-5.

31. Abrams, B.L., F.H. Alsikafi, and N.G. Waterman, Colostomy: a new

look at morbidity and mortality. Am Surg, 1979. 45(7): p. 462-464.

32. Guenaga, K.F., et al., Ileostomy or colostomy for temporary

decompression of colorectal anastomosis. Cochrane Database Syst

Rev, 2007(1): p. CD004647.

33. Ellis, H., H. Gajraj, and C.D. George, Incisional hernias: when do they

occur? Br J Surg, 1983. 70(5): p. 290-1.

34. Scarpa, M., M. Barollo, and M.R. Keighley, Ileostomy for constipation:

long-term postoperative outcome. Colorectal Dis, 2005. 7(3): p. 224-7.

51

35. Burgess, P., V.V. Matthew, and H.B. Devlin, A review of terminal

colostomy complications following abdominoperineal resection for

carcinoma. Br J Surg, 1984. 71: p. 1004.

36. Eldrup, J., et al., Parakolostomihernier. Incidens og relation till

stomiens placering. Ugeskr Laeger, 1982. 144: p. 3742-3743.

37. Ortiz, H., et al., Does the frequency of paracolostomy hernias depend

on the position of the colostomy in the abdominal wall? Int J

Colorectal Dis, 1994. 9(2): p. 65-7.

38. Londono-Schimmer, E.E., A.P. Leong, and R.K. Phillips, Life table

analysis of stomal complications following colostomy. Dis Colon

Rectum, 1994. 37(9): p. 916-20.

39. Leong, A.P., E.E. Londono-Schimmer, and R.K. Phillips, Life-table

analysis of stomal complications following ileostomy. Br J Surg, 1994.

81(5): p. 727-9.

40. Birnbaum, W. and P. Ferrier, Complications of abdominal colostomy.

Am J Surg, 1952. 83: p. 64-67.

41. Cevese, P.G., et al., Peristomal hernia following end-colostomy: a

conservative approach. Ital J Surg Sci, 1984. 14(3): p. 207-209.

42. Cheung, M.T., Complications of an abdominal stoma: an analysis of

322 stomas. Aust N Z J Surg, 1995. 65(11): p. 808-11.

43. Pearl, R.K., et al., Early local complications from intestinal stomas.

Arch Surg, 1985. 120(10): p. 1145-1147.

44. Sjodahl, R., B. Anderberg, and T. Bolin, Parastomal hernia in relation

to site of the abdominal stoma. Br J Surg, 1988. 75(4): p. 339-41.

45. Baslev, A., Kolostomitilvaerelse. Ugeskr Laeger, 1973. 135: p. 2799-

2804.

52

46. Stelzner, S., G. Hellmich, and K. Ludwig, Die versorgong der

Parakolostomiehernie nach Sugarbaker. Zentralblatt für Chirurgi,

1999. 124(Suppl 2): p. 13-17.

47. Everingham, L., The parastomal hernia dilemma. World Council of

Enterostomal Therapists Journal, 1998. 18: p. 32-34.

48. Tretbar, L., Kirurgi vid stomikomplikationer. Stomijournalen: nordisk

tidskrift för stomi vård, 1988. 2(4): p. 10-11.

49. Caricato, M., et al., Retrospective analysis of long-term defunctioning

stoma complications after colorectal surgery. Colorectal Dis, 2007.

9(6): p. 559-61.

50. Gogenur, I., et al., Prevention of parastomal hernia by placement of a

polypropylene mesh at the primary operation. Dis Colon Rectum,

2006. 49(8): p. 1131-5.

51. Robertson, I., et al., Prospective analysis of stoma-related

complications. Colorectal Dis, 2005. 7(3): p. 279-85.

52. Hammond, T.M., et al., Parastomal hernia prevention using a novel

collagen implant: a randomised controlled phase 1 study. Hernia,

2008. 12(5): p. 475-81.

53. Berger, D., Prevention of parastomal hernias by prophylactic use of a

specially designed intraperitoneal onlay mesh (Dynamesh IPST).

Hernia, 2008. 12(3): p. 243-6.

54. Singh, G., J.M. Wilkinson, and D.G. Thomas, Supravesical diversion

for incontinence: a long-term follow-up. Br J Urol, 1997. 79(3): p. 348-

53.

55. Weaver, R.M., J. Alexander-Williams, and M.R. Keighley, Indications

and outcome of reoperation for ileostomy complications in

inflammatory bowel disease. Int J Colorectal Dis, 1988. 3(1): p. 38-42.

53

56. Carlsen, E. and A. Bergan, Technical aspects and complications of

end-ileostomies. World J Surg, 1995. 19(4): p. 632-6.

57. Marshall, F.F., W.F. Leadbetter, and S.P. Dretler, Ileal conduit

parastomal hernias. J Urol, 1975. 114(1): p. 40-2.

58. Edwards, D.P., et al., Stoma-related complications are more frequent

after transverse colostomy than loop ileostomy: a prospective

randomized clinical trial. Br J Surg, 2001. 88(3): p. 360-3.

59. Tilney, H.S., et al., Comparison of outcomes following ileostomy

versus colostomy for defunctioning colorectal anastomoses. World J

Surg, 2007. 31(5): p. 1142-51.

60. Ho, K.M. and D.P. Fawcett, Parastomal hernia repair using the lateral

approach. BJU Int, 2004. 94(4): p. 598-602.

61. Bloom, D.A., H.B. Grossman, and J.W. Konnak, Stomal construction

and reconstruction. Urol Clin North Am, 1986. 13(2): p. 275-83.

62. Fontaine, E., et al., Twenty-year experience with jejunal conduits.

Urology, 1997. 50(2): p. 207-13.

63. Farnham, S.B. and M.S. Cookson, Surgical complications of urinary

diversion. World J Urol, 2004. 22(3): p. 157-67.

64. Wood, D.N., et al., Stomal complications of ileal conduits are

significantly higher when formed in women with intractable urinary

incontinence. J Urol, 2004. 172(6 Pt 1): p. 2300-3.

65. Hulten, L., J. Kewenter, and N.G. Kock, [Complications of ileostomy

and colostomy and their treatment]. Chirurg, 1976. 47(1): p. 16-21.

66. Pearl, R.K., et al., A survey of technical considerations in the

construction of intestinal stomas. Annals of Surgery, 1988. 51: p. 462-

465.

67. Todd, I.P., Intestinal stomas. 1978, London: Heinemann Medical

Books.

54

68. Goligher, J.C., Extraperitoneal colostomy or ileostomy. Br J Surg,

1958. 46: p. 97.

69. Devlin, H., Management of of Abdominal Hernias. 1988, London:

Butterworths.

70. Martin, L. and G. Foster, Parastomal hernia. Ann R Coll Surg Engl,

1996. 78(2): p. 81-4.

71. Marks, C.G. and J.K. Ritchie, The complications of synchronous

combined excision for adenocarcinoma of the rectum at St Mark's

Hospital. Br J Surg, 1975. 62(11): p. 901-905.

72. Stephenson, B.M., et al., Minimal anatomical disruption in stoma

formation: the lateral rectus abdominis positioned stoma (LRAPS).

Colorectal Dis, 2009.

73. de Ruiter, P. and A.B. Bijnen, Successful local repair of

paracolostomy hernia with a newly developed prosthetic device. Int J

Colorectal Dis, 1992. 7(3): p. 132-4.

74. Leslie, D., The parastomal hernia. Surg Clin North Am, 1984. 64(2): p.

407-15.

75. De Raet, J., et al., Waist circumference is an independent risk factor

for the development of parastomal hernia after permanent colostomy.

Dis Colon Rectum, 2008. 51(12): p. 1806-9.

76. Carne, P.W., et al., Parastomal hernia following minimally invasive

stoma formation. ANZ J Surg, 2003. 73(10): p. 843-5.

77. Liu, J., et al., Stoma formation for fecal diversion: a plea for the

laparoscopic approach. Tech Coloproctol, 2005. 9(1): p. 9-14;

discussion 14.

78. Hamada, M., et al., Laparoscopic permanent sigmoid stoma creation

through the extraperitoneal route. Surg Laparosc Endosc Percutan

Tech, 2008. 18(5): p. 483-5.

55

79. Resnick, S., New method of bowel stoma formation. Am J Surg, 1986.

152(5): p. 545-8.

80. Anderson, I.D., et al., An improved means of faecal diversion: the

trephine stoma. Br J Surg, 1992. 79(10): p. 1080-1.

81. Rubin, M.S., D.J. Schoetz, Jr., and J.B. Matthews, Parastomal hernia.

Is stoma relocation superior to fascial repair? Arch Surg, 1994. 129(4):

p. 413-8; discussion 418-9.

82. Rieger, N., et al., Parastomal hernia repair. Colorectal Dis, 2004. 6(3):

p. 203-5.

83. Cheung, M.T., N.H. Chia, and W.Y. Chiu, Surgical treatment of

parastomal hernia complicating sigmoid colostomies. Dis Colon

Rectum, 2001. 44(2): p. 266-70.

84. Allen-Mersh, T.G. and J.P. Thomson, Surgical treatment of colostomy

complications. Br J Surg, 1988. 75(5): p. 416-418.

85. Horgan, K. and L.E. Hughes, Para-ileostomy hernia: failure of a local

repair technique. Br J Surg, 1986. 73(6): p. 439-440.

86. Baig, M.K., et al., Outcome of parastomal hernia repair with and

without midline laparotomy. Tech Coloproctol, 2006. 10(4): p. 282-6.

87. Pearl, R.K. and J.H. Sone, Management of peristomal hernia:

techniques of repair, in Nyhus and Condon´s Hernia, R.J. Fitzgibbons

and A.G. Greenburg, Editors. 2002, Lippincott Williams & Wilkins:

Philadelphia. p. 415-422.

88. Kasperk, R., U. Klinge, and V. Schumpelick, The repair of large

parastomal hernias using a midline approach and a prosthetic mesh in

the sublay position. Am J Surg, 2000. 179(3): p. 186-8.

89. Cassar, K. and A. Munro, Surgical treatment of incisional hernia. Br J

Surg, 2002. 89: p. 534-545.

56

90. Schumpelick, V. and U. Klinge, Incisional abdominal hernia: the open

mesh repair. Langenbecks Arch Surg, 2004. 389: p. 1-5.

91. Stephenson, B.M. and R.K. Phillips, Parastomal hernia: local resiting

and mesh repair. Br J Surg, 1995. 82(10): p. 1395-6.

92. Kish, K.J., B.R. Buinewicz, and J.B. Morris, Acellular dermal matrix

(AlloDerm): new material in the repair of stoma site hernias. Am Surg,

2005. 71(12): p. 1047-50.

93. Kelly, M.E. and S.W. Behrman, The safety and efficacy of prosthetic

hernia repair in clean-contaminated and contaminated wounds. Am

Surg, 2002. 68(6): p. 524-8; discussion 528-9.

94. Geisler, D.J., et al., Safety and outcome of use of nonabsorbable

mesh for repair of fascial defects in the presence of open bowel. Dis

Colon Rectum, 2003. 46(8): p. 1118-23.

95. Aldridge, A.J. and J.N. Simson, Erosion and perforation of colon by

synthetic mesh in a recurrent paracolostomy hernia. Hernia, 2001.

5(2): p. 110-2.

96. Morris-Stiff, G. and L.E. Hughes, The continuing challenge of

parastomal hernia: failure of a novel polypropylene mesh repair. Ann

R Coll Surg Engl, 1998. 80(3): p. 184-7.

97. Israelsson, L.A., Preventing and treating parastomal hernia. World J

Surg, 2005. 29(8): p. 1086-9.

98. Rosin, J.D. and R.A. Bonardi, Paracolostomy hernia repair with Marlex

mesh: a new technique. Dis Colon Rectum, 1977. 20(4): p. 299-302.

99. Bayer, I., S. Kyzer, and C. Chaimoff, A new approach to primary

strengthening of colostomy with Marlex mesh to prevent

paracolostomy hernia. Surg Gynecol Obstet, 1986. 163(6): p. 579-

580.

57

100. Light, H.G., A secure end colostomy technique. Surg Gynecol Obstet,

1992. 174(1): p. 67-68.

101. Schumpelick, V., et al., Minimized polypropylene meshes for

preperitoneal mesh plasty in incisional hernia. Chirurg, 1999. 70: p.

422-430.

102. Olsen, M., et al., Assessment of accuracy and reliability in

preoperative templating for hip resurfacing arthroplasty. J

Arthroplasty, 2010. 25(3): p. 445-9.

103. MKAPPASC.SPS., Available from

ftp://ftp.spss.com/pub/spss/statistics/nicols/macros/mkappasc.sps;2008

104. Bringman, S., et al., Hernia repair: the search for ideal meshes.

Hernia, 2010. 14(1): p. 81-7.

105. Schumpelick, V., Meshes: Benefits and Risks. 2004: p. 91-179.

106. Helgstrand, F., I. Gogenur, and J. Rosenberg, Prevention of

parastomal hernia by the placement of a mesh at the primary

operation. Hernia, 2008. 12(6): p. 577-82.

107. Janes, A., Y. Cengiz, and L.A. Israelsson, Experiences with a

Prophylactic Mesh in 93 Consecutive Ostomies. World J Surg, 2010.

34: p. 1637-1640

108. Abcarian, H., Peristomal hernias, ed. S.D. Wexner and A.M. Vernava.

1995: Igaku-Shoin: New York. 449-452.

109. Cengiz, Y. and L.A. Israelsson, Parastomal hernia. European Surgery,

2003. 35: p. 28-31.

110. Janson, A., A. Jänes, and L.A. Israelsson, Laparoscopic stoma

formation with a prophylactic prosthetic mesh. Hernia, 2010. 14: p.

495-498.

58

59

ACKNOWLEDGEMENTS

I wish to express my gratitude and appreciation to all those who have helped

me completing this thesis, with special thanks to

Leif Israelsson, my tutur, for scientific knowledge, inspiration, impressive

patience, years of support and friendship in research, clinical work and life.

Yucel Cengiz and Daniel Millbourn, for their support and friendship.

Rosemarie Christmansson and Ann-Christin Lindström, for ideas and help

in conducting the studies.

Märth-Marie Berglund, without whom nothing would work.

Thomas Alexander Halliday, for swift and skilful linguistic revision of text

being at times only remotely connected to English.

Lena Weisby, for knowledgable co-authorship and energy.

60

The Research and Development Unit, Västernorrland, for providing time

and financial support.

The Hudiksvall musketeers, friends for life.

Friends and colleagues at the Department of Surgery, Sundsvall, for their

support and for allowing me to combine scientific and clinical work.

My tutors and colleagues at the Department of Surgery, Hudiksvall

Hospital, for giving me an extraordinary start in the practice of surgery.

Tönu and Margareta, my parents, and Peter, my brother, for your love and

support in life.

Elin, for everlasting support, incredible patience and for being the love of my

life.

Joel, Jakob, Simon, my joy and pride, giving meaning to life.