Immunological Studies in the Postsplenectomy Syndrome

By I. Claret, L. Morales, and A. Montaner

K I N G A N D S C H U M A C H E R ~ in 1952 were the first authors to point out the serious effects of infections in splenectomized children. It was later

confirmed by a number of workers whose work has been compiled by Singer. 2 The present report is an a t tempt to relate the seriousness of the postsplenec- tomy infectious processes to failures in the immune mechanisms of the spleen.

For a long time it was thought that the spleen, although part of the reticulo- andothelial system, was not of essential importance to the organism and could be removed by surgery with impunity. The overwhelming infection which can occur after removal of the spleen, especially in young children, has indicated the need for a more detailed and deeper study of the splenic immune functions. These abnormal infections in children may depend upon a transitory alteration of the immune mechanism, or upon an alteration of the phagocytes, both mobile and fixed, in the reticuloendothelial system. Reports of work carried out in the study of serial levels of immunoglobul ins in splenectomized children are contradictory. Huang et al. 3 a n d R o w l e y 4 found variations in the humoral response after splenectomy, while Salaw 5 and Thurman 6 found no variation in asplenic children. The reduction in Ig M has been pointed out by King and Schumacher I to be responsible for a higher degree of susceptibility to infec- tions, as it is known that the organism responds to an antigenic stimulus initially with macroglobulin (Ig M) antibodies and not until later with Ig A and Ig G. Hobbs 7 has pointed out that a reduction of Ig M increases susceptibility to infections with meningococci, an organism which in all reports of neopost- splenectomy infections shares the high infection rate with the pneumococcus. By contrast, an increase in the lg A was observed by Wasi s after splenectomy in thalassemic patients. This increase in thalassemia may be caused by in- creased synthesis in other lymphoid organs or may be due to a reduced catabo- lism. Hepat ic lesions, according to Janeway, 9 are accompanied by hypergam- maglobul inemia with a high Ig A, which splenectomy may increase even further. It is also possible that the spleen intervenes in the distribution of Ig A and thus provokes its increase through a double mechanism.

MATERIALS AND METHODS

We have done a detailed immunological study among 52 splenectomized children during the period 1962-1964 and followed them for from 4 mo to I1 yr after operation. The reason for splenectomy was portal hypertension in six children, splenic rupture in three, thrombocytopenic

From the Department of Pediatric Surgery, Division of Pediatrics, University of Barcelona Faculty of Medicine, Barcelona, Spain.

Presented before the XXlst Annual Congress of the British Association of Paediatric Surgeons. Berne, Switzerland August 27-31, 1974.

Address for reprint requests: Professor 1. Claret, Departments Cirugia Pediatrica, Facultad de Medicina, Hospital Clinico, Barcelona, Spain.

�9 1975 by Grune & Stratton, Inc.

Journal of Pediatric Surgery, Vol. 10, No. 1 (February), 1975 5g

60 CLARET, MORALES, AND MONTANER

Table 1. Splenectemy Related to Diagnosis and Age

Age (yr)

Splenectomy I 1 2 3 4 5 6 7 8 9 10 11 Total

Thalassemia 1 2 3 2 1 1 1 11

Spherocytosis 1 1 2 1 2 1 3 2 1 1 1 16

Portal hypertension 1 1 2 1 1 6

Thrombocytopenia 1 2 2 3 2 1 1 1 13

Trauma 1 1 1 3

Aplastic anemia 1 1 Enzymopenic anemia 1 1

Incidental splenectomy 1 1 Total cases 2 2 3 2 7 8 10 6 3 5 1 3 52

purpura in 13, aplastic anemia in one, enzymopenic anemia in one, hereditary spherocytosis in 16, and removal of spleen incidental to other surgery in one (Table 1). The results are compared between two groups of children: one control group composed of 30 children between 3 and 10 yr who underwent appendectomy and were followed for 2 yr, recording when serious infections oc- curred. A second control group, composed of 76 children who had no previous immunological defect or repeated common infections or isolated serious infections whom we had followed after different surgical or orthopedic procedures. Both control groups were followed for 2 yr carrying out a scheduled immunologic study.

The immunological control was centered in the study of serial immunoglobulin and of globulin /3-1 metal binding. The determining of Ig A, lg M, and/3-1 metal binding has been carried out with Behringwerke "'Tripartigen" discs in agar gel according to plain radial immunodiffusion technique described by Mancini and later confirmed by Carbonara and Heremans. Ig A, lg M, and globulin /3-1 metal-binding immunoglobulins were given to the splenectomized group before operation and at the time of follow-up (4 me to 5 yr). We carried out the same program with 30 children with appendectomy before operation and 2 yr later. This test was carried out to determine whether the remOval of the appendix, a lymphoid organ of secondary importance, produced im- munological alterations, as Laski and McMillan I~ found (four cases of meningitis among 94 appendectomized children). A period of 2 yr was considered to be the time of highest risk of serious infections in splenectomized children.

The immunologic data have been submitted to statistical analysis by means of the Student test in a Compucorp 145 E Statistician (ATATO) of the Barcelona Faculty of Medicine. The changes in each of the immunoglobulins and the globulin /3-1 metal binding have been evaluated sta- tistically over a period of 2 yr in all groups of children.

RESULTS

In the g roup of 76 n o r m a l children only two had serious infections; none

died. The immuno log i ca l s tudy carr ied out init ially and 2 yr later showed that:

Ig A increases r emarkab ly (p = 0.0005), which we cons ider to be related to

the chi ld 's growth; the ages o f this con t ro l g roup var ied be tween 2 and 12 yr.

E ighteen were, on first observa t ion , under 5 yr, by which t ime the Ig A has

no t reached adul t values. The Ig A and the globul in/3-1 meta l b inding showed

no r emarkab l e changes. Ig M showed only a slight increase (p = 0.025).

In the g roup of 30 appendec tomized chi ldren dur ing the per iod o f 2 yr there

were no deaths and only one case o f serious infection. The values ob ta ined in

these chi ldren f rom the two de te rmina t ions o f i m m u n o g l o b u l i n and globul in

/%1 meta l b inding showed a significant increase o f the Ig G (p = 0.0005). The

grea t difference in age (2.5 to 12.5) detracts f rom the va lue o f these data be-

cause a lmos t 50~o of the chi ldren are 5 or less than 5 yr old. The Ig A and the

POSTSPLENECTOMY SYNDROME 61

n cases

cured inf.

exitus

t halassemia !!!!!!! i

spherocytosis

thrombocytopen ia

porta l hypertension

ap las t i c anemia

O 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

n c a s e s

Fig. 1. Morbidity and mortality in postsplenectomy children related to diagnosis.

globulin/3-1 metal binding gave the same result as in the preceding controls. The values obtained at the second determination when compared with the val- ues obtained at the second determination in normal children was not significant in any of the parameters measured.

In splenectomized children the mortality was 11~. Two of the children died from their disease (portal hypertension and thalassemia), four children (7.35~) died from infection. There were five other serious infections that were success- fully treated; one child with thalassemia presented with pneumococcic pneu- monia on two occasions, 10 and 17 mo after splenectomy. Serious infections, then, resulted in death (Fig. 1).

The etiologic agent in these infections is shown in Table 2. The predominat- ing organism was pneumococcus, with a 40~o mortality, a far higher percentage than that of the pneumococcal infections in nonsplenectomized children. In children with a previous disorder of the reticuloendothelial system the differ- ence in the percentage of serious infections (with and without healing) are quite high. Children with a previous reticuloendothelial disorder have a mortality six times greater than those who do not (36.11~o vs 6.4~o).

The immunologic determinations demonstrated a very significant increase of the lg A (p = 0.001) (Fig. 2A), a little significance to the increase of the lg G (p = 0 .0 l ) (F ig . 2B), a scarcely significant reduction of the IgM (p = 0.001) (Fig. 2C), and no significant changes of the globin B-I metal binding (Fig. 2D).

Table 2. Etiology of Severe Infections in 52 Postsplenectomy Children

Portal Aplastic Infecting Thrombocytopenia Spherocytosis Tha lassemia Hypertension Anemia

Organism Cure Died Cure Died Cure Died Cure Died Cure Died

Pneurnococcus 1 2 1 1

Meningococcus 1

Staphylococcus 1 Pseudornonas 1

Unknown 1

4 0 0 "

3 0 0

2 0 0 .

100"

A

I | I �9

! 2 3 4

years 6f e v o l u t i o n

150

100

1 2 3 4

C y e a r s o f e v o l u t i o n

See l e g e n d on f a c i n g p a g e .

2000

1500

1000

5O

! i | J 0 2 3 4 5

B y e a r s o f e v o l u t i o n

500

4 0 0 '

3 0 0

2 0 0 �84

f

100

D years of evolution

POSTSPLENECTOMY SYNDROME 63

Comparing the values found in splenectomized children (second determina- tion) with 76 children belonging to the normal group, Ig A was higher in splenectomized children (p = 0.001) and Ig M was lower and also significant (p -- 0.001).

Comparison of values found in splenectomized children with the values of the group of 30 appendectomized children showed the following: a reduction in Ig M and an increase in the Ig A in the splenectomized children, both sig- nificant (p = 0.001) while there was an increase of Ig G of little significance (p --- 0.025). Globulin/3-1 metal binding, as in the preceding comparison, re- vealed no difference. A comparison of the values found at first and second determination in each group of splenectomized children shows the following:

Thalassemia: Ig G underwent a slight nonsignificant increase (p = 0.05) as did Ig A (p = 0.0025). Ig M decreased (p = 0.001).

Spherocytosis: Ig G was essentially unchanged. Ig A was increased, and Ig M was decreased, each significantly (p -- 0.001).

Portal hypertension: Ig G and Ig A showed no significant increase (p = 0.01); Ig M did decrease significantly (p = 0.005).

Thrombocytopenia: the parameters which have been studied present the same variations as in spherocytosis.

The globulin /3-1 metal-binding values showed no important variations in any of these diagnoses. Comparing the splenectomized children with the normal children, the most outstanding was the following: a continuous decline in the Ig M value which was observed in all the cases with statistical significance and a rise in Ig G in thalassemia and portal hypertension (p = 0.001 in the former and p = 0.0025 in the latter). In the remaining groups Ig G does not change significantly.

We can make two subgroups in the splenectomized children, those with a previous disorder of the reticuloendothelial system and those without. Compar- ing the values, we found that Ig G in children with a previous disorder of the reticuloendothelial system had higher values (p = 0.001).

A separate comparison of these two groups with the parameters found in normal children and appendectomized children gives the following: In both groups, with and without previous reticuloendothelial system disorders, the val- ues of Ig A are higher and those of Ig M are significantly decreased. The Ig G in those presenting a previous reticuloendothelial disorder are significantly higher (p = 0.0005). In those without a previous disorder, Ig G showsaao im- portant variation. The immunological studies carried out have shown a con- stant decline in all Ig M values. This decline should be related to the produc- tion of antibodies of a macroglobulin type against the infecting organisms, especially those with a polysaccharides capsule.

SUMMARY

Variations in the serial immunoglobulins of 52 children splenectomized for a variety of indications were studied and compared with two groups of

Fig. 2. (A) Ig A levels in postsplenectomy syndrome; (B) Ig G levels in postsplenectomy syn- drome; (C) Ig M levels in postsplenectomy syndrome; (D) transferase levels in postsplenectomy syndrome.

64 CLARET, MORALES, AND MONTANER

chi ldren , one p o s t o p e r a t i v e and one w i thou t o p e r a t i o n . The mos t s ignif icant f inding was the c o n s t a n t decrease in Ig M in every instance. O u r series of sp l enec tomized ch i ld ren seems to conf i rm the facts no t ed by o the r au tho r s . The inc idence o f ser ious infec t ions has been c lear ly s ignif icant and is c lear ly post - s p l e n e c t o m y sepsis. The age o f the pa t i en t s has a g rea t influence on the ser ious- ness o f the sept ic processes , these be ing m o r e severe on younge r ch i ldren . The inf luence has also been revealed o f the cause which m o t i v a t e d the ope ra t i on , this c o r r e s p o n d i n g in the first p lace to tha l a s semia and in the second p lace t h r o u g h sphe rocy tos i s and p s e u d o h o r m o n e s .

I t is ev ident tha t the i m m u n o l o g i c role o f the spleen dur ing ear ly c h i l dhood , especia l ly under the age o f 5 yr is i m p o r t a n t . The grea tes t risk o f pos t sp lenec- t o m y sepsis is in the first 2 yr o f life. S p l e n e c t o m y shou ld be de l ayed unti l a f te r the age o f 5 if c l in ica l c i r cums tances permi t . I f not , p r o p h y l a c t i c chemo- t h e r a p y shou ld be ca r r i ed out dur ing the first 18 mo af ter sp lenec tomy.

REFERENCES

1. King H, Schumaker MB: Splenic studies, 1. Susceptibility to infection after splenectomy performed in infancy. Ann Surg 136:239, 1952

2. Singer DB: Postsplenectomy sepsis. Per- spect Pediatr Pathol 1:285, 1973

3. Huang NN, Sheng KT, Pilling GP: Anti- body response, to Vi antigen administered sub- cutaneously and intravenously following sple- nectomy in children. Am J Dis Child 100:699, 1960

4. Rowley DA: The formation of circulating antibody in the splenectomized human being following intravenous injection of heterologous erythrocytes. J Immunol 65:515, 1950

5. Salaw S, Bouroncle BA, Wall RL, et al:

Studies on the antibody response in splenec- tomized persons. N Engl J Med 261:120, 1959

6. Thurman WG: Splenectomy and immu- nity. Am J Dis Child 105:138, 1963

7. Hobbs JR, Milner RDG, Watt PJ: Gamma-M deficiency predisposing to menin- gococe'al septicaemia. Br Med J 4:543, 1967

8. Wasi P: Streptococcal infection leading to cardiac and renal involvement in thalas- semia. Lancet 1:949, 1971

9. Janeway ChA: The immunological system of the child. Pa Immunology deficiency. Arch Dis Child 41:366, 1966

10. Lasky B, MacMillan A: Incidence of in- fection in children after splenectomy. Pediatrics 24:253, 1959