tubular and tubular necrosis in thekidneys ofburned … · "acute tubular necrosis." the...

J. clin. Path. (1956), 9, 279.

DISTAL TUBULAR AND PROXIMAL TUBULAR NECROSISIN THE KIDNEYS OF BURNED PATIENTS

BY

S. SEVITTFrom the Pathology Department, Birmingham Accident Hospital, and

M.R.C. Burns Research Unit

(RECEIVED FOR PUBLICATION SEPTEMBER 29, 1955)

The histological pictures associated with acuterenal failure following abortion, incompatibleblood transfusion, crush injuries, severe trauma,blackwater fever, sulphonamide intoxication, andmany other conditions including burns have beendescribed by various writers (Bratton, 1941 ; Dunn,Gillespie, and Niven, 1941; Bywaters and Beall,1941; Bywaters and Dible, 1942; Erb, Morgan,and Farmer, 1943; Shen, Ham, and Fleming,1943; Lucke, 1946; Goodpastor, Levenson,Tagnon, Lund, and Taylor, 1946; Oliver, Mac-Dowell, and Tracy, 1951; Oliver, 1953; Bull andDible, 1953). Lucke, whose report included 48fatally burned subjects, claimed that there was acommon morphological picture for acute renalfailure caused by a variety of conditions. Hecoined the term "lower nephron nephrosis" be-cause he considered the essential lesion to be aselective focal degeneration or necrosis of the thickHenle and distal convoluted tubules. This namehas been criticized by Oliver and by Dible (Bulland Dible, 1953) because they believe that a wide-spread tubular necrosis is the essential feature,and they renamed the morphological pattern" acute tubular necrosis."The present report is based on a histological

analysis of the kidneys of 86 fatally burnedpatients. Some of these patients had a necrosisof the proximal convoluted tubules whilst othersshowed a histological picture similar to lowernephron nephrosis. In this paper, the former con-

dition will be termed "acute proximal tubularnecrosis " (P.T.N.) and the latter " acute distaltubular necrosis" (D.T.N.).

Proximal and distal tubular necrosis in burnedpatients are not identical phenomena. Proximaltubular necrosis occurs mainly in elderly subjectswho have nephrosclerosis and is usually associatedwith a severe oliguria. Distal tubular necrosismainly affects children and younger adults andmay or may not be associated with acute renal

2E

failure. If the necrosis affects many nephrons andis widespread through the kidneys it is termed" diffuse D.T.N." and this is associated withanuria or severe oliguria or alternatively with anon-oliguric form of acute uraemia (Sevitt, 1956).If the condition affects few nephrons it is termed" focal D.T.N." and this is infrequently associatedwith acute renal failure.

THE PATIENTSDuring the six and half years from 1948 to the

middle of 1955, 174 burned patients (109 females)died in the Birmingham Accident Hospital and 90necropsies were performed. For various reasonsfour are excluded and the present analysis is basedon the remaining 86 subjects. Forty-four of thesewere female. The ages varied from 16 months to87 years: 21 were less than 5 years old, 16 werebetween 5 and 15 years, 11 between 15 and 40years, 14 were between 40 and 60 years, and 24were older than 60 years (Table I).

TABLE IAGE ANALYSIS

No. of Age Group in YearsSubjects .With < 5 S-15 15-40 40-60 >60

Distal tubular nec-rosis.. .. 13 9 4 4 4

Focal .. 9 6 2 1 0Diffuse .. 4 3 2 3 4

Proximal tubularnecrosis 0 0 2 4 11

Notubularnecrosis 8 7 5 6 9

Totals . 21 16 11 14 24

Most of the patients were extensively burned.In 74 subjects the burns involved 20% to 90% ofthe body area (20% to 90% burns); 29 patientshad 20 to 40% burns, 24 between 40% and60%, and in 21 patients the burns involved morethan 60% of the body surface (Table II). In only

copyright. on S

eptember 4, 2020 by guest. P

rotected byhttp://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.9.4.279 on 1 Novem

ber 1956. Dow

nloaded from

http://jcp.bmj.com/

S. SEV1TT

TABLE IIEXTENT OF BURNS

No. of Burn Y. of Body AreaSubjectsWith < 20 20-40 40-60 >60

Distal tubular nec-rosis . .. 0 4 13 17

Proximal tubular nec-rosis . 4 9 2 2

Early P.T.N. 3 6 2 2No tubular necrosis 9 15 9 2

Totals 12 29 24 21

eight patients were the burns less than 10% andseven of them were over 70 years of age.The survival periods varied from two hours to

four and half months; 10 subjects died within 24hours of burning, 10 within one to two days, 20within two to five days, 17 within five to 14 days,and 29 lived longer than 14 days.

Seventy-nine of the 86 necropsies were per-formed within 36 hours of death, 63 within 24hours, and 30 within 12 hours.

ROUTINE TREATMENTDuring the first one to three days-the shock

stage-80 patients were transfused with plasma;the other six patients were elderly subjects withrelatively small burns. Transfusion was begunsoon after admission and usually within a fewhours of burning. The degree of haemoconcen-tration on admission and during the course oftransfusion was estimated by serial haematocritobservations from which the amount of plasmalost from the circulation was calculated. In recentcases blood volume studies have been carried outby Dr. E. Topley. These tests guided the amountand speed of plasma transfusion through whicholigaemia was mainly combated. Some patientswere also transfused with blood at this stage but inothers blood transfusion was delayed for a weekor longer. Other intravenous fluids (dextran 6%,glucose 5%, saline 0.9%) were also given to somepatients. During the shock stage, the total volumesof intravenous fluids given were usually between3 and 10 litres but varied from 700 ml. to 18.6litres in the different patients according to the age,extent of burn, and other factors. Oral fluids,such as glucose-water, sweetened fruit juice, orsodium lactate, were given when vomiting did notprohibit their use. Two or three days later, feed-ing by an egg-milk mixture was begun and wascontinued until an adequate diet could be taken10-14 days after burning. Skin grafting wasusually carried out after two or three weeks andblood transfusion was given during each opera-

tion. Some patients were grafted during the firstfew days.The burns were first dressed at the end of the

shock phase when penicillin cream and/or otherlocal antibiotics were applied to prevent or con-trol wound infection. Systemic therapy withpenicillin, aureomycin, polymyxin, or erythro-mycin, or a combination was introduced when in-fective complications occurred.

Biochemical analysis of the blood and urineguided the giving of supplementary sodiumchloride, potassium salts, etc.

HISTOLOGICAL METHODSOne or more blocks from each kidney were

examined. Paraffin sections from both kidneys wereroutinely stained by haematoxylin and eosin and withthe picromallory technique. Sections containing castswere stained by Pickworth's benzidine method andsome of these kidneys were also examined unstained.Other staining methods when appropriate includedphosphotungstic-acid-haematoxylin, Weigert's fibrin,Prussian blue, periodic-acid-Schiff, and phloxin-tartrazine techniques. Frozen sections were preparedand stained for fat by Sudan IV or oil red 0.

CLASSIFICATIONOn a histological basis the kidneys were classi-

fied as follows:

Distal tubular necrosis ...Proximal tubular necrosisPyelonephritis ... ...

Subacute glomerulo-nephritis ... ...

No significant abnor-mality .. ... ...

Total ... ...

No. of Patients34 (diffuse 16, focal 18)172

3286

In general, distal and proximal tubular necrosiswere morphologically distinct but there were a fewsubjects in whom both conditions were found (twoclassified as D.T.N. and three or more as P.T.N.).The cases of pyelonephritis and glomerulonephritiswill receive no further attention. Those classifiedas having no significant abnormality showed noevidence of distal or proximal tubular necrosis butincluded cases with cloudy tubular swelling.Table I lists the number of patients who had

distal tubular necrosis, proximal tubular necrosis,and those without tubular necrosis in the agegroups <5, 5 to 15, 15 to 40, 40 to 60, and > 60years. Only D.T.N. occurred in children under15 years (22 cases) and most of those with tubularnecrosis who were over 60 years had P.T.N. (11out of 15 patients). Proximal tubular necrosis didnot occur in children and became progressivelymore frequent with advancing age, whilst the fre-

2.80

copyright. on S



j.com/

J Clin P


ber 1956. Dow

nloaded from

http://jcp.bmj.com/

RENAL TUBULAR NECROSIS AFTER BURNING

quency of diffuse D.T.N. was about the same inall age groups. When the number of cases ofD.T.N. and P.T.N. in each age group are addedtogether and the sum is expressed as a percentageof the total number of subjects in the age group,the frequencies of tubular necrosis in the differentage groups are about the same. They are 62%(13 out of 21 patients under 5 years), 56% (5 to 15years), 54% (15 to 40 years), 57% (40 to 60 years),and 62% (over 60 years). In other words, thetotal incidence of tubular necrosis is the sameirrespective of age and about 50% to 60% of allfatally burned subjects at all ages develop one orother form of tubular necrosis. Age, however,may determine which kind of tubular necrosis thepatient may suffer. In children this will be a focalor diffuse D.T.N., in adult life either diffuse D.T.N.or proximal tubular necrosis may occur but withadvancing age P.T.N. becomes the usual kind.This indicates either that there is some factor inelderly subjects which is not present in youngerpzople and which facilitates the development ofP.T.N., or alternatively that younger patientspossess some property which facilitates the de-velopment of D.T.N. The former appears morelikely and renal arteriosclerosis may be the pre-disposing factor (vide infra).

DISTAL TUBULAR NECROSISDistal tubular necrosis is classified into diffuse

and focal forms according to whether many orfew nephrons are affected. The terms refer to thehistological density of lesions, but in both diffuseand focal kinds the individual tubules are com-monly affected in a discrete or focal fashion. Thehistological picture also varies with the period ofsurvival, and early and later stages are recognized.Diffuse D.T.N. was found in the kidneys of 16patients (nine adults, seven children, 10 females)who died 12 hours to 23 days after burns involv-ing 37% to 80% of the body area. The meanarea burned was 67%. Focal D.T.N. was foundin 18 patients (10 females), 15 of whom werechildren. They died 15 hours to 35 days afterburns involving 27% to 90% of the body area.The mean area burned was 62 %.

Diffuse D.T.N.: Early StageThis appeared in the kidneys of eight patients

who survived 12 hours to two and half days.Severe oliguria rapidly developed in five of themeven though transfusion with plasma appeared tobe adequate. Their total excretion of urine variedonly from 90 to 196 ml. during the 30 hours totwo and half days of survival. The urine flow of

two others was adequate, but the blood urea ofone (age 5 years) was 80 mg.% 12 hours beforedeath which suggested that uraemia might havedeveloped had the patient survived longer. Thesenon-oliguric patients are described elsewhere(Sevitt, 1956). Haemoglobinuria was reported insix patients.Gross Appearance.-The kidneys of seven

patients were moderately or deeply congested andsome dripped blood from the cut surfaces. Twowere a purplish, cyanosed colour.

Histology.-Histologically the main feature wasthe presence of numerous haemoglobin and eosino-philic granular casts in many collecting (Fig. 1) andwide Henle tubules (Fig. 2) and often in secondconvoluted tubules. The casts often distended thetubules and appeared to block them. Many ofthem were amorphous fused masses of haemo-globin, others were coarsely or finely granular.Occasional groups of red cells were seen withintubules, but these were not a feature. Many of theeosinophilic granular casts which usually stainedred or pink in the picromallory preparation werealso benzidine positive. Albuminous exudate wasfrequently mixed with the casts. Early tubulo-epithelial necrosis was common in tubules dis-tended by casts, was usually focal and was mani-fest by nuclear pyknosis and by local thinningand eosinophilic staining of the cytoplasm (Fig. 2).Focal necrosis of Henle tubules which did not con-tain casts was not common but appeared as agranular cytoplasmic disintegration with nuclearkaryolysis. In one patient many focal areas ofnecrosis affected the proximal convoluted tubules,that is, both distal and proximal tubular necrosiswere present. The glomerular capsules generallycontained many coarse eosinophilic granules aboutthe size of erythrocytes, also albuminous exudateand occasionally a few red cells. Irregulardesquamation of the parietal layer of Bowman'scapsules was frequent. The hyperaemia noted inthe gross specimens was histologically confirmed,but irregular small foci of cortical ischaemia werealso seen and affected some glomerular tufts andthe related peritubular capillary network. In onlyone case was relative ischaemia a feature. Thrombiin glomerular capillaries were seen in threepatients.

Diffuse D.T.N.: Later StageA more advanced histological pattern was ob-

served in the kidneys of eight patients. Clinicallythey are divided into two groups. The first is offour patients who rapidly became severely oliguricor anuric, then uraemia developed and consider-

281

copyright. on S



j.com/

J Clin P


ber 1956. Dow

nloaded from

http://jcp.bmj.com/

< P _ L 5 L 8ij- * able azotaemia with hypocalcaemia, hyperpotass-aemia, hyperphosphataemia, hypochloraemia, andhyponatraemia were found. They died between

+ §^ u* .,; 9' four and five days after burning. For example, aV ' , man of 24 years with 75% burns passed 168 and

,gf>;*=, S)* * v / ¢ tt jiX 20 ml. of urine respectively during the first two#PA/W days, failed to pass any more urine and died five_tF*^I*4 s 8^/*t*^ days after burning with a blood urea of 284 mg.

,A^**t + ^ ¢#> ; X **. %. Similarly a girl of 9 years with 55% burnspassed 130, 137, 69, 5, and 4 ml. of urine on suc-

*4>;vss/_t I£cessive days. Her blood urea level rose from 159mg. % on the third day to 318 mg. % on the fifth

' - day when she died, her serum potassium andphosphate values rose to 37 mg. % and 9.2 mg. %

t-t¶7;*Xrespectively, and her serum sodium and calcium*l .S '<^ ^ ,levels fell to 278 mg. % and 7.7 mg. %. Early4V_!f; -s - haemoglobinuria was reported in each patient.

X as^2These patients represent the classical anuric-£i + -< > 4 'fO, uraemic form of acute renal failure.

The second group consists of three patients who" survived six, 10, and 21 days, and who also had

uraemia with high blood urea levels. However,their daily urinary output was considered to be

A *S * * 3W -t_adequate and oliguria did not occur (Sevitt, 1956)..mb*̂|>*2- <#_ Haemoglobinuria had been reported in two of

r*$: * w ! x ***X¢e} them. The eighth patient, a baby of 18 months,survived 23 days but neither her urinary output

* s . w*/>fnor her blood urea levels are known.FIG. 1

Fis. I and 2.-Early diffuse D.T.N.in a patient with 75% burns who 14:survived two days. In Fig. 1 A*many collecting tubules in thepyramid are plugged withgranular haemoglobin castsW(haematoxylin and eosin,x 210), Aand in Fig. 2 many Henle _t Jtubules contain haemoglobincasts and show focal tubular w _ rnecrosis at the sites of blockage(haematoxylin and eosin, x WV380). .;

FIG. 2 4b D.E ...% , 4 ,...,.

copyright. on S



j.com/

J Clin P


ber 1956. Dow

nloaded from

http://jcp.bmj.com/

FIGS. 3 and 4.-Diffuse D.T.N. in aNt patient with 80% burns who died

on the tenth day. A vein in theboundary zone is thrombosed andshows early organization. Migra-tion of tubular epithelium aroundpart of the clot indicates tubulo-venous anastomosis. (Haematoxylinand eosin, x 220.) In Fig. 4 thereis a lymphoid cell infiltration in thecortex related to a ruptured second

parenchymawssrtconvoluted tubule. (Haematoxylinand eosin, x 220.)

patcual atthiaesoheprmis w3v f

4~~~~~ ~ ~ ~ ~~~~~~~~~~l

The kidneys were swollen and enlarged and ,,th3 capsules were under tension the cortices^00w%' *7were widened, the outlines were hazy, and .sometimes they were pale. In some cases the -parenchyma was softer than usual and in four *patients minute greyish-white foci werevisible,Wparticularly at the bases of the pyramids. c

Histologically the characteristic changes arecasts and tubular necroses in the distal and collect- VAing tubules, tubulo-e ithelial regeneration, tubulo-venous anastomoses and venous thrombi (Fig.3), interstitial oedema and an inflammatoryicellular exudate in the boundary zone, aroundivenules or related to extruded casts or rupturedstraight or convoluted distal tubules (Fig. 4).For example, in one of the patients who died \ ;

with uraemia and anuria,,numerous coarse and 6finely granular haemoglobin and eosinophiliccasts4were found mainly in the collecting tubules and a f_ascending limbs of Henle, and some casts alsocontained desquamated epithelial cells. Some l *.colloid or albuminous casts were also seen. Dis-crete necroses of straight tubules, particularly inthe boundary zone and in the pyramid, werepresent both at the sites of blockage and else

and regeneration of thin epithelium had begunand had already produced some peculiar tubularforms. Interstitial oedema and collections oflymphocytes and some plasma cells were seen L

copyright. on S



j.com/

J Clin P


ber 1956. Dow

nloaded from

http://jcp.bmj.com/

S. SEVITT

around and between many straight and convolutedtubules, particularly those ruptured or blocked bycasts. Lymphoid cells were also found aroundsmall venules in the boundary zone and elsewhere.Some tubulo-venous anastomoses produced pre-sumably by rupture were seen. The proximal con-voluted tubules showed evidence of cloudy swell-ing and their lumina contained albuminousexudate. Bowman's capsules also contained someexudate. Some glomeruli in this patient and inone other of this group contained capillarythrombi.

In those who survived a longer period epithelialregeneration and the interstitial inflammatorychanges were generally more advanced. Forexample, in one patient who died 10 days afterburning, tubular reparative efforts had producedmany plicated and bizarre tubular forms. Numer-ous haemoglobin, cellular and other casts, andfocal necroses at the sites of blocked tubules werevisible. Numbers of the venules in the boundaryzone were thrombosed and parts of the thrombiwere surrounded by a single layer of tubularepithelium which had spread from a nearby rup-tured and regenerating tubule (Fig. 3). Focalcollections mainly of lymphoid and plasma cells,with a few eosinophil leucocytes and oedema ofthe interstitial tissue, were present. They wereseen between and around straight tubules andvenules particularly in the boundary zone andwere also related to second convoluted tubules(Fig. 4). The parietal layer of many Bowman'scapsules had become cubical or even low columnaras if over-vigorous replacement of previouslydesquamated epithelium had occurred. Somecloudy swelling was found in the proximaltubules; the cortex and medulla were congestedand the latter was oedematous.

In one patient haemoglobin casts were notfound and haemoglobinuria was not reported: inanother case pigmented casts were few. Instead,eosinophilic colloid and granular casts which werebenzidine-negative were relatively numerous in thedistal and collecting tubules but otherwise thehistological appearances differed in no way fromthe other cases.A common, almost universal feature in the early

and later stages of diffuse D.T.N. was the presenceof many unusual mononuclear cells within venules,particularly in the pyramids. As Dible (Bull andDible, 1953) stated, they appeared similar to manybone marrow cells, and he suggested that they arosefrom the vascular endothelium and were the resultof stagnation of the venous blood flow. Theappearance was seen both in the oliguric and non-

*1A0.0 ~

FIG. 5.-Focal D.T.N. in a patient with 65%. burns who survivedsix days. An occasional Henle tubule is distended with ahaemoglobin cast and shows focal pykno-necrosis of the epi-thelium. (Haematoxylin and eosin, x 210.)

oliguric forms. Indeed, no histological differencesbetween these two clinical varieties of diffuseD.T.N. could be made out.

Focal D.T.N.: Early StagesThis was observed in four patients who sur-

vived 15 hours to two and half days. Three wereknown to have had haemoglobinuria. Severeoliguria rapidly developld in one of them; in twoothers the urinary flow was adequate or evenpolyuric and in the fourth patient it was notknown.

Histologically only a few scattered haemoglobincasts-amorphous and granular eosinophilic forms-were present in a few collecting, ascending Henle(Fig. 5) and second convoluted tubules. Thenumber of casts varied in different parts of thesections but were rarely more than 1 or 2 per low-power field and were often less frequent. Themajority of tubules did not contain casts andappeared to be normal. Early discrete tubularnecroses (pyknosis) were restricted to some of theblocked tubules, but in one case focal necroses ofa few wide Henle tubules which did not contain

284

copyright. on S



j.com/

J Clin P


ber 1956. Dow

nloaded from

http://jcp.bmj.com/


wcasts were also seen. Extrusion ofcasts from ruptured tubules wasseen in the anuric case. Numbersof glomerular capillary thrombiwere found in one patient.

Focal D.T.N.: Later Changes *The kidneys of the 14 patients in E

this group are classified histo-logically into three subgroups:"active," "healing," and "healed"types. The "active" type num-bered six patients, five of whomdied two and a quarter to eight and ,a half days after burning and theother survived 35 days. In this sub-ject the onset of the disorder may * thave started days or weeks afterburning. Oliguria developed after36 hours in the child who survived * ,.only two and a quarter days. Inall the others the urinary flow was *inormal, but one developed uraemia eand this was attributed, at least inpart, to the severe proximal tubulardegeneration which was also present f(Table IV). The histologically"healing " and " healed " groups pconsisted of eight children, none ofwhom was oliguric. One child died focalcollecti.only seven days after burning and Recent epithethe others survived 15 to 26 days. y e

Altogether eight patients with focal D.T.N.(early and later stages) who were biochemicallyinvestigated showed no evidence of azotaemia.Six others were not investigated.

Histologically the " active " form was character-ized by the presence of a few haemoglobin andeosinophilic casts in the distal convoluted, ascend-ing Henle (Fig. 5) and collecting tubules in fourpatients and by colloid casts in a fifth. Theepithelium of the distended parts was usually thin,a few localized necrotic epithelial foci were seenand tubular regeneration was in progress particu-larly in the patient who survived 35 days. Focalcollections, mainly of lymphoid cells, were foundaround some of these tubules (Fig. 6) or aroundthe walls of veins in the boundary zone of thecortex or in both situations. Occasionally thesevenules were thrombosed and organization of thethrombi was in progress (Fig. 7). Essentially thehistological picture was similar to that in thediffuse form except that few tubules were involved.The sixth patient differed from the others in

that multiple haemoglobin casts were present, but

.T.N. in a patient with 45% burns who survived 35 days. An occasionaljDn of lymphoid cells surrounds a Henle tubule blocked by a cast.lial regeneration has occurred. Boundary zone of cortex. (Haema-:osin. x 380.)

few tubules appeared to be blocked and only occa-sional pykno-necrotic epithelial foci were seen:in other words a diffuse cast state was associatedwith focal tubular necrosis. Had the patient sur-vived and most of the casts been removed by theadequate urine flow, a characteristic picture offocal D.T.N. would have remained.The morphological changes which are believed

to represent the " healing " stage or the " healed "

residue of a recent focal D.T.N. consist of focalcollections mainly of lymphoid cells and occasionalorganizing venous thrombi. The former werelocated around the walls of small veins, particu-larly in the boundary zone of the cortex, or wererelated to second convoluted tubules or werepresent in both situations. The organizing venousthrombi were also usually seen in the boundaryzone and probably represented former tubulo-venous rupture. Direct evidence of tubularnecrosis was found in only two subjects, and inthree cases a few granular or haemoglobin castswere seen in distal tubules. In four patients partsof an occasional straight Henle tubule showed

285

copyright. on S



j.com/

J Clin P


ber 1956. Dow

nloaded from

http://jcp.bmj.com/

S. SEVITT

FIG. 7.-Focal D.T.N. in a patient with 65% burns who survived six dalof an occasional vein in the boundary zone. Tubular epithelia]indicates tubulo-venous anastomosis. (Haematoxylin and eosin, x

evidence of recent regeneration. The term " heal-ing" is applied when tubular evidence of necrosisor regeneration still persisted: the kidneys of fourpatients were so classified. The term " healed "is used when there was no tubular evidence offocal tubular necrosis. There were five such casesincluding one case classified as proximal tubularnecrosis. In these the characteristically locatedlymphoid cell collections, and in one patient anorganizing venous thrombus, were accepted as theresidue of a focal distal tubular necrosis. Theinterstitial cellular collections might have led to a

diagnosis of interstitial nephritis if the entire histo-logical process was not considered. It may bethat some cases diagnosed histologically as inter-stitial nephritis because of focal collections oflymphoid cells are in reality the healed or healingresidue of distal tubular necrosis.

ACUTE PROXIMAL TUBULAR NECROSIS

Proximal tubular necrosis was found in 17patients, excluding the two with distal tubularnecrosis in whom severe proximal tubular degener-

^ ation was also present. Histo-logically the condition was

[^[ v : $ characterized by a widespreadk 4' focal or a confluent necrosis

W°* 9 § particularly of the proximalconvoluted tubules, whilst fewif any of the tubules in the

:- pyramid were involved.,* Unlike those with distal

tubular necrosis, all theset patients were adults and most* of them were elderly (Table

I) and had nephrosclerosis.e Thirteen of them were 58 to

87 years old and eight wereK8* over 70 years. Twelve hadw moderate or severe nephro-

sclerosis due to renal arterio-.8~d sclerosis of senile or other

form,- and five had renal andcardiac evidence of hyper-tension. Ten were females.

^*> The extent of burning wasusually less than in those with

^^_* ;$} distal tubular necrosis; ninepatients had burns between20% and 40%, four were less

Saff8extensively burned, and inL.* only four were the burns

ys. Venous thrombosis more extensive (Table II).I migration into the clot:210.) The patients are divided

into two groups, those whodied within five days of burning, the 13 earlycases, and those who survived 10 days or longer,the four later cases. This division was made be-cause in at least eight of the early cases acuterenal failure had occurred. Seven had anuria orsevere oliguria and another (Case 8, Table III)had incipient uraemia without oliguria. His bloodurea level rose from 80 to 190 mg. %. Case 2,for example, passed no urine during his 18 hoursof survival, Case 3 passed 218 ml. in 1 1/6 days,Case 7 passed 570 and 55 ml., and Case 10 passed1,115, 170, and 130 ml. respectively on successivedays. Four oliguric patients were known to havesignificant azotaemia with blood urea levels of 70,85, 110, 140 mg. % respectively (Table III). Theacute renal failure was probably related to theburns and to the associated histological picture ofproximal tubular necrosis.On the other hand, none of the later cases

developed oliguria and the histological tubularnecrosis was not known to be associated withclinical effects. In these patients the necrosis wasprobably not directly related to the burning but

286

copyright. on S



j.com/

J Clin P


ber 1956. Dow

nloaded from

http://jcp.bmj.com/


TABLE IIIPATIENTS WITH ACUTE PROXIMAL TUBULAR NECROSIS

I~~~~a80°) 0 Other1V204)

0.~o ~' ~ Features

243M18 ,, ~~~~Oi+ 70+ PlmnrE,arly Deaths1 87 F 2 hr. + Coronary and

(801/.) cerebralatheroma

2 43 M 18 ,, 0 + 70 + Pulmonary(60%) oedema

3 75M Ilkdays + + ++(400%)

4 87 F 1*, + ?+ + Coronary(45%) atheroma

5 65 M lj ++ + 85(25%)

6 72 F 2 + + Diabetes. Cere-(15%) bral atheroma

7 47 M 2 ,, 0 + 145 + Glycosuria(70%)

8 20 M 21 ,, 0 0 or 80- + "Healed" focal(35%) slight 190 D.T.N. also

seen. Acutehaemolyticjaundice

9 80 F 3 ,, ++ ?+ Cerebral ather-(25%) oma. Early

broncho-pneumonia

10 58 F 3 + + 110 + Overtransfusion,(35%) oedema + +

II1 67 F 3 ,, + + Coronary(25%) atheroma

12 71 F 5 + ?0(10%/1)

13 7SF 5 ,, +

Later Deaths14 70 F 10 , + ?0 Cerebral infarc-

(6%) tion. Broncho-pneumonia

15 28 M 12 0 0 Broncho-(35%) pneumonia

16 63 F 17 ,, 0 0 Septicaemia,(25%) pneumonia

17 58 M 70 ,, + 0 Hepaticjaundice,(25%) pulmonary

embolism

was possibly associated with infective complica-tions (bronchopneumonia, septicaemia) in Cases14 to 16 and with hepatitis and jaundice in Case 17.

The KidneysCortical pallor was present in six early cases

but the kidneys of one (Case 8) were very con-gested. In two subjects a number of pale, moreor less circumscribed cortical areas were sur-rounded by narrow red borders (Fig. 8). Thesepale zones were restricted to the outer two-thirdsor so of the cortex and the deeper cortex andmedulla were congested. This appearance sug-gested a cortical ischaemia and medullary by-passof the type described by Trueta, Barclay, Daniel,Franklin, and Prichard (1947).Cortex-The esseintial change was a widespread

necrosis of the proximal convoluted tubules(Figs. 9 and 11). Their nuclei had either dis-appeared or stained feebly as if in process of solu-

I .1 t . 1 7 " it 12 1

FIG. 8.-Left kidney, Case 2, 60% burns, survived 18 hours. Partsofthe cortex are pale and these areas are surrounded by a narrowred border giving an infarct-like appearance.

tion. In most patients the tubular epithelial cellswere loosened from each other and from the base-ment membrane. In some places they had losttheir tubular arrangement, the lumen had dis-appeared, and the cells had become more or lessjumbled and irregular but were restricted by theperitubular membrane (Fig. 11). In other casesthe cells were only slightly loosened and weremostly still arranged around a central lumen (Fig.9). The fine peritubular membrane was alwaysintact. The extent of proximal tubular necrosisvaried but was considerable in each case. In fiveof the 13 early cases (Cases 1, 2, 6, 12, and 13)necrosis involved most of or all the proximaltubules (+ + + necrosis); in six others (Cases 3,5, 7, 8, 10, and 11) about half the tubules wereobviously necrotic, whilst the remainder appearedviable (+ + necrosis); and in the remaining two(Cases 4 and 9) the necrosis was focal and lessextensive ( + necrosis). In Case 3 there was + +necrosis in one kidney and a + necrosis in theother. In three of the later cases necrosis was+ + + but in Case 17 it was + +.Necrosis of the second convoluted tubules

was also present but was less extensive (Fig. 9).Histologically viable but collapsed spiral tubuleswith stainable nuclei were often surrounded bynecrotic proximal tubules. In the cortex manyof the wide straight tubules were necrotic, par-ticularly the proximal descending parts of thefirst tubules but also many ascending Henle limbs.The narrow descending limbs and collectingtubules usually appeared less affected or even

287

copyright. on S



j.com/

J Clin P


ber 1956. Dow

nloaded from

http://jcp.bmj.com/

S. SEVITT

: .......* 0y.j*i,, .t * .................Sf.igt.-+-* e4.4W."~~~~~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 4

t~~~~Nisg,6 ' / ' 9. *; 15 |-

"C~~~~~~~~~~~4.,SX*a4*Xf:4t)O*isE*s*p,..tA4v IF; sw*t

,.-Ass 4 . i2, * * a; ,t* S 4,iitS Sr¢ ,P;R ' $t*<,A~~~~~~~~~~~~.* A

FIG. 9FIGS. 9 and 10.-Proximal tubular necroais, early case, 70°/ burns, survived two days (Case 7). In Fig. 9 the proximal

convoluted tubules are obviously necrotic but their general architecture is still intact. The second convoluted tubulesare less affected. In Fig. 10 the tubules in the pyramid are separated by oedema, but are otherwise normal. (Haematoxylinand eosin, x 210.)

viable. The glomerular tufts appeared viable inall except three cases. In Case 6 all glomeruli andtubules were obviously necrotic (Fig. 11) and smallhaemorrhages were present. In Cases 2 and 12 afew of the glomerular tufts were undergoingnecrotic changes whilst the majority appearedviable.

Subcapular Viability.-Subcapsular rims orzones of histologically viable cortex were seen in13 cases. The rims were usually thin, irregularand often interrupted by foci of necrotic tubules,but in two cases the viable zone extended into theouter quarter or so of the cortex.

Vascular Changes.-In 13 cases there was his-tological evidence of cortical ischaemia. In three(Cases 6, 12, and 13) most of the peritubularcapillary networks were empty of blood, and inthe others the ischaemic zones were focal andirregular. A variable number of glomerular tuftswere also empty but only in Cases 2 and 3 wasthis a prominent feature.

Multiple focal haemorrhages were seen in fivesubjects; in Cases 6, 7, and 13 they were related

to dilated cortical veins, in Case 12 they werefocal, capillary peritubular haemorrhages, and inCase 4 they occurred in the subcapsular zone.

In Cases 3 and 4 many glomerular capillarythrombi were seen. Arterial thrombosis was foundin Case 2 (vide infra).

Pyramid.-In the boundary zone a variablenumber of Henle tubules were obviously necrotic,but in the pyramids the architecture and nuclearstructure of the collecting and Henle tubules werenormal (Fig. 10). They were usually collapsedand empty and separated by oedema. In Case 4there was a hyaline droplet degeneration of theepithelium (in this case there was also consider-able exudation into the glomerular spaces). In 10cases the vasae recti were congested but in fiveothers these vessels were empty.

Casts and Disal Tubular Necrosis.-In 15 casescasts were present. Mostly they were of the kindcommonly found in nephrosclerotic kidneys, butin seven cases (Cases 2, 3, 5, 7, 8, 9, and 12) atleast some of them contained haemoglobin andthey were mostly found within second convoluted

288

copyright. on S



j.com/

J Clin P


ber 1956. Dow

nloaded from

http://jcp.bmj.com/

RENAL TUBULAR NECROSIS AFTER BURNING:F 7 f~~i t ri ° . ->e1

FIG. 11.-Proximal tubular necrosis, early case, 15% burns, survived two days (Case 6). Advancednecrosis of the cortical tubules, also glomerular necrosis. ,Note the group of hyalinized arterioles.(Haematoxylin and eosin, x 210.)

and collecting tubules. Such casts were notnumerous, but in Cases 2 and 3 (possibly alsoCases 1 and 12) the appearance was suggestive ofearly distal tubular necrosis.Other Features.-In Case 8 the presence of focal

collections of lymphoid cells around venules par-ticularly in the boundary zone suggested the"healed " residue of a previous focal D.T.N. un-related to the burning.

Nephrosclerotic atrophy and focal fibrosis ofthe cortex was seen in 12 cases: in five a hyalinechange affected some or many of the afferentglomerular arterioles.One case merits a fuller description.Case 2.-A man, aged 43 years, suffered 60%

burns from an oxy-acetylene explosion and died 18hours later. Plasma transfusion was begun one hourafter burning and 3.5 litres was given. Haemo-concentration was inadequately combated because thehaematocrit rose from 45% to 55% on at least twooccasions but fell again. His peripheral circulationwas poor; he became restless, violent, and irrationaland then comatose. No urine was passed and theblood urea level at death was 70 mg.%.

Necropsy (22 hours after death) revealed viscidblood, swelling of the brain, and oedema of thetongue, laryngeal opening, and of both lungs. Thebladder contained a few millilitres of bloody urine.The kidneys together weighed 131 oz., were swollen,and their capsules were under tension. When thesewere stripped the surface of the right kidney wasuniformly congested but a number of pale, obviouslyischaemic areas were seen on the left cortex (Fig. 8).These varied in shape and size, were 0.5 cm. to severalcentimetres long, were well demarcated by a narrow(1 mm. diameter) red border and contrasted with theremainder of the cortex, which was congested. Alto-gether about one-sixth to one-eighth of the cortex wasinvolved in these pallid zones. Longitudinal sectionshowed that the pallor was restricted to the outer two-thirds or three-quarters of the cortex and did notinvolve the juxtamedullary cortex or the pyramidswhich were dark red.Histology.-The pale areas in the left kidney were

due to ischaemia of the glomeruli and of the capillarynetwork, whilst the surrounding cortex, much of thejuxtamedullary cortex, and the pyramids werehyperaemic. In addition there was an irregular alter-nation of ischaemic and hyperaemic cortical foci ofmicroscopic size in the other parts of the left kidney

289

copyright. on S



j.com/

J Clin P


ber 1956. Dow

nloaded from

http://jcp.bmj.com/

S. SEVITT

and throughout the right organ but in the latterischaemic foci were more numerous. Groups of emptyglomerular tufts and peritubular capillaries alternatedwith hyperaemic groups. The epithelium of mostof the proximal and distal convuluted tubules wasundergoing obvious necrotic changes or severedegenerative changes approaching a necrosis. Thiswas more extensive in the right kidney and moreadvanced in some areas than in others. Similarchanges involved many of the wide Henle limbs inthe boundary zone but the descending limbs were lessaffected. The lumina of the affected tubules containedalbuminous and granular debris, and in parts of thedistal tubules much desquamated necrotic epitheliumwas also present. The peritubular basement mem-branes were intact. Most of the glomerular tufts ap-peared viable, but a few also were undergoing necrosis.Under the capsule was a narrow irregular rim ofhistologically viable tubules. Haemoglobin and othercasts were seen in numbers of distal and collectingtubules particularly in the right kidney. Some of themwere focally obstructed and necrotic suggesting anearly distal tubular necrosis. Otherwise the tubulesof the pyramid were collapsed and separated byoedema but appeared histologically normal. Throm-bosis of arcuate, interlobar, and some interlobulararteries was seen in the left kidney mainly in relation-ship to the ischaemic cortical zones. Glomerularcapillary thrombosis was not found. Nephroscleroticchanges were slight.The morbid anatomy and histology suggest that the

proximal tubular necrosis was the result of corticalischaemia. Whilst a part of this may have been inthe form of a cortical by-pass mechanism similar tothat described by Trueta et al. (1947), most of theischaemia was not of this kind but was focal, irregu-lar, and widespread and related to many separategroups of afferent arterioles. The thrombosed arterieswere more likely the result rather than the cause ofthe cortical ischaemic zones. This case is similar tothat described by Brown and Crane (1943), but theirpatient survived several days and necrosis had alsoinvolved many glomeruli.

Glomerular Capylay ThrombosisThis condition was seen in the glomeruli of

eight subjects. Histologically the thrombi, whichoften blocked and distended many of the capil-laries of a single tuft, usually appeared as granu-lar eosinophilic elongated masses taking the shapeof the capillary lumen. They were stained red inthe picromallory preparation and dark blue orblack by the phosphotungstic-acid-haematoxylintechnique. Sometimes the thrombus or part ofit was relatively wide and then it grossly distendedpart of the capillary loop. Thrombi were notseen in the afferent arterioles. In four patientshalf the glomeruli or more were involved but inthe others fewer tufts contained thrombi.

Five of the eight patients were over 70 yearsof age, the others were also adults and the condi-tion was not seen in children. Five patients hadnephrosclerosis and the thrombosis involved thecapillaries of both normal and abnormal glomeruli.The condition was found in five out of 10 patients(50%) who died between 24 and 48 hours afterburning and who showed either proximal ordistal tubular necrosis. It was seen in only oneof the seven subjects who died within 24 hoursof burning with distal or proximal tubularnecrosis, in only two of the 17 such patients whosurvived two to five days after burning, and innone of those who survived longer. Thus throm-bosis of the glomerular capillaries was particu-larly found among the elderly people who diedwith distal or proximal tubular necrosis duringthe second day after burning. Its aetiology isnot clear. The thrombi may be formed locallywithin glomerular capillaries, the endothelium ofwhich has become temporarily abnormal. Alter-natively they may be partly embolic in origin andresult from the trapping of fragmented erythro-cytes and other microscopic debris within theglomerular capillaries. However they are formed,they appear to be a temporary phenomenon sincethey are rarely seen in patients who survive morethan two days. It is unlikely that they play asignificant part in the pathogenesis of tubularnecrosis. The condition was known to Wertheim(1867) but has not been described by recentauthors.

DISCUSSIONBy histological analysis the kidneys of 86 burned

patients included 34 subjects with distal tubularnecrosis and 17 with proximal tubular necrosis.These are histologically distinct although therewere a few cases in which there was evidence ofboth conditions.

Distal tubular necrosis occurred both in a diffuseform when many nephrons were affected (16 cases)and in a focal form when few nephrons were in-volved (18 cases). Diffuse D.T.N. was similar tothe appearance described in burned patients byPack (1926), Shen et al. (1943), Gibson (1944), andGoodpastor et al. (1946). Essentially it had thesame morphological pattern as Lucke's lowernephron nephrosis. Histologically it was charac-terized by the early appearance of haemoglobincasts in many collecting and wide Henle tubulesand in distal convoluted tubules. Frequenttubular blockage by casts was often associatedwith local epithelial necrosis which occurred lessfrequently in tubules which did not contain casts.Rupture of tubules, dislocation of casts, tubular

290

copyright. on S



j.com/

J Clin P


ber 1956. Dow

nloaded from

http://jcp.bmj.com/


TABLE IVFREQUENCY OF HAEMOGLOBINURIA IN PATIENTSWITH AND WITHOUT TUBULAR NECROSIS AND IN

PATIENTS WITH RENAL FAILURE*

Distal Proximalt NoTubular Necrosis Tubular Tubular

Necrosis NecrosisDiffuse Focal

All cases .. .. 12/16 6/18 6/13 3/24Anuria or severe

oliguria .. 8/9 112 4!7 0l 17/24Uraemiawithout oli- (?5/9) oS (?18/26

* Numerals are no. of patients with haemoglobinurialtotal no. inthe group.

t Excludes four who died more than 10 days after burning (seeTable III).

$ One patient with incipient uraemia.§ Also severe proximal tubular degeneration.

regeneration, interstitial oedema, peritubular andperivenous infiltration by lymphoid and othercells, tubulo-venous anastomoses, and venousthrombi were found in patients who survivedsufficiently long. Clinically the condition wasassociated with actual renal failure which tookone of two courses, oliguric or non-oliguric (TableIV). Nine subjects rapidly developed severeoliguria or anuria and four of them survivedsufficiently long to develop uraemia. The urineflow of five others was more or less normal, andif oliguria occurred it was slight and transient:four of these subjects were known to have uraemiain one of whom it was incipient. Thus acuteuraemia may occur with or without oliguria.Clinically the non-oliguric form may be over-looked but is important to recognize since recentadvances in therapy (high-calorie low-nitrogenfeeding) may increase the patient's chance of sur-vival. Both Shen et al. (1943) and Goodpastor etal. (1946) refer to burned patients with significantazotaemia but without oliguria, the kidneys ofwhom had changes now referred to as distal tubu-lar necrosis. Acute uraemia without oliguria isnot restricted to burned patients. Two severelyinjured patients studied by Burnett were possiblyexamples (Smith, 1951), and more recentlyTeschan et al. (1955) described cases of uraemiawithout oliguria occurring after battle wounds.One case of post-traumatic uraemia withoutoliguria in which the histology was that of a diffuseD.T.N. was described by the writer (Sevitt, 1956).Most of the 18 patients with focal D.T.N. were

children (Table I). The urinary excretion of 15of them was known to be adequate and only twodeveloped severe oliguria (Table IV). The causesof death were unrelated to renal failure and wereusually infective conditions like bronchopneu-monia or septicaemia. One patient developed

uraemia without oliguria, but this may have beenthe result of a severe proximal tubular degenera-tion: Eight others were biochemically investi-gated and showed no evidence of significantazotaemia. Histologically the early phases of thefocal condition differed in no qualitative way fromthe diffuse form; the difference was quantitative.Casts were few, necroses were not numerous, fewtubules were involved. The involvement of fewnephrons was the essential- histological differencefrom the diffuse form. The diagnosis of the stageof healing or the healed residue was made largelyby the finding of characteristically situated cellularfoci around venules or between tubules in theboundary zone or at second convoluted tubules:in some cases this was associated with thromboticevidence of tubulo-venous rupture. In the heal-ing stage there was also residual evidence oftubular necroses or recent regeneration.The 17 patients who died with extensive proxi-

mal tubular necrosis were divided into 13 earlycases dying within five days of burning and fourlater cases surviving 10 days or longer. Eight ofthe former were known to have acute renal failurewhich took the form of severe oliguria or anuriain seven and incipient uraemia without oliguriain the other. Most of the patients were elderlyand the majority had nephrosclerosis with orwithout hypertension.The overall frequency of P.T.N. in the present

series was about 20% (17 among 86 subjects). Thisis probably an underestimation of its true inci-dence among the 174 fatally burned patients, be-cause necropsies were performed on 55% to 73%of patients in various age groups below 60 yearsbut only in 32% of the 76 patients over the age of60 years.The histological appearance was characterized

by a widespread focal or confluent necrosis of theproximal convoluted tubules and the condition wasrestricted to the cortex. Post-mortem autolysiswas not responsible for the histological changes,first because the necropsies were not unduly de-layed (six were performed within 15 hours ofdeath, 10 within 24 hours, and 16 within 36 hours);secondly, in most patients there was a histologic-ally viable subcapsular rim or area of cortex;and thirdly the pyramid was not necrotic. Thehistological appearance had many features incommon with that described by Dunn and Mont-gomery (1941) in eight unburned patients underthe name acute necrotizing glomerulo-nephritis.Five of their patients were known to have hadsevere oliguria or anuria and there was a notablerise in the blood urea in the three patients on

291

copyright. on S



j.com/

J Clin P


ber 1956. Dow

nloaded from

http://jcp.bmj.com/

S. SEVITT

whom the test was done. They noted the rapidityof nuclear loss and cytoplasmic changes in theaffected epithelium, and this has been observed inthe present series.

PafhogenesisProximal Tubular Necrosis.-The renal capsule

has a separate blood supply and anastomoses withthe main renal vessels in the subcapsular cortex.Thus the presence of a viable subcapsular rimof cortex indicated that the tubular necrosisof the cortex was the result of ischaemic changesin the blood supply from the main renal vessels.The histological viability of the tubules in thepyramids could have been the result either of theirgreater resistance to ischaemic anoxia, or alter-natively to an abnormal intrarenal distribution ofthe blood flow by means of which cortical but notmedullary ischaemia is produced. de Wardener(1955) completely occluded the main left renalartery of eight dogs for three to four hours, thenreleased the vessels and killed the dogs a few dayslater. In each left kidney there was considerableor complete necrosis of the cortex (often includingglomeruli) but in several dogs there were thinirregular subcapsular areas of viable cortex. Theleast affected areas of the cortices in all thenecrosed kidneys were in the juxtamedullary zone.On the other hand the tubules of the pyramid wereviable in appearance in five dogs and only in threedid necrosis occur here. de Wardener's sectionshave been examined by me and his results con-firmed. Thus the distal and collecting tubules ofdogs frequently withstand complete ischaemia last-ing three or four hours whilst the proximal tubulesusually suffer irreversible necrotic changes duringthis period. In man it is therefore likely that aperiod of total renal ischaemia sufficiently long toproduce proximal tubular necrosis may be in-sufficient to produce serious changes in the distalor collecting tubules. The viable appearance ofthe distal and collecting tubules in the presentseries of patients with cortical tubular necrosis isconsistent with this view.However, an abnormal distribution of the intra-

renal blood supply either immediately after burn-ing or after a period of total ischaemia may alsohave occurred. There was histological evidenceof cortical ischaemia in 13 cases, but in most ofthese the ischaemic zones were focal, multiple, andof irregular distribution. This may have been pro-duced by narrowing of the afferent glomerulararterioles when both the glomeruli and peritubularcapillaries were empty of blood, or by narrowingof the efferent arterioles when the former werecongested and the latter empty. These processes

have been shown to occur in some experimentallyburned rabbits, the intrarenal circulation of whichhas been studied by the injection in vivo of a col-loidal gold suspension into the abdominal aortain a cephalic direction (Sevitt, unpublished obser-vation). In two patients, however, the appearanceof cortical, infarct-like areas in the kidneys indi-cated that ischaemia of these areas was the resultof a cortical by-pass mechanism. Trueta et al.(1947) showed that in this by-pass the blood forthe distal two-thirds or so of the cortex is divertedthrough the juxtamedullary glomeruli, the afferentvessels of which loop downwards and draindirectly into the tributaries of the renal vein.Moreover they showed that this is likely to happenin nephrosclerotic kidneys because occlusion ofthe capillaries of the juxtamedullary glomerulioften results in a direct communication betweenthe afferent and efferent arterioles. This hasrelevance to the present cases most of which wereelderly patients with nephrosclerosis.

Studies of the intrarenal blood flow in severelyburned animals have been made by Monsaingeon,Tanret, and Daussy (1949) by the injection ofvarious dyes in vivo. They found ischaemia ofthe outer two-thirds of the cortex in 14 out of 28burned rabbits, in three out of nine burned guinea-pigs, and in five out of 25 burned rats. Total renalischaemia was found in four rabbits, four guinea-pigs, and eight rats, and irregular cortical ischaemiawas found in seven rabbits and six rats. This fre-quent and extensive ischaemia was not found bythe writer in burned rabbits (unpublished observa-tions), but the animals of Monsaingeon and hiscolleagues were usually burned more extensivelyand for longer periods than the writer's.From a review of the literature of cortical

necrosis in unburned patients Dunn and Mont-gomery came to the conclusion that infectionsplayed a part in the genesis of many cases includ-ing five of their own. An infective aetiology mayaccount for the changes in three or four of thelater cases of the present series but not for theearly cases. Renal necrosis in these patients musthave been produced by changes set in motion bythe burning of the skin which resulted, as hasalready been indicated, in complete renal ischaemiaor by a disturbance of intrarenal circulation or inthe latter follow ng a period of complete ischaemia.Case 1, however, may be an exception since shedied only two hours after burning and she mayalready have had proximal tubular necrosis at thetime she was burned.

Renal or intrarenal ischaemia is probably pro-duced by vasoconstriction but the mechanism

292

copyright. on S



j.com/

J Clin P


ber 1956. Dow

nloaded from

http://jcp.bmj.com/

RENAL TUBULAR NECR OSIS AFTER BURNING

whereby this occurs is not known with certainty.Renal vasoconstriction, perhaps reflexly produced,may result from the oligaemia due to loss ofplasma into the burned skin. Oligaemia, however,is usually more severe and of more rapid onsetin more extensively burned patients, but only fourof the subjects with proximal tubular necrosis hadburns exceeding 40% of the body area (Table II).rhis may be compared with an incidence of 30out of the 34 patients with distal tubular necrosis.Nervous or humoral factors have been incrimin-ated and there is evidence of these in experiment-ally burned animals. Barac (1946, 1948) found thatthe anuria in burned dogs was partly determinedthrough excitation of the renal nerves which wasboth of cerebral and of spinal origin, and Page(1943) showed that the plasma of burned dogsdeveloped vasoconstrictor properties for thevessels of isolated rabbits ears.

Evidently further information is required butthe main body of evidence indicates that proximaltubular necrosis after burning is produced by renalischaemia of vasomotor origin.

Diffuse D.T.N.-Three hypotheses have beenput forward to explain the severe oliguria andanuria of the classical type of acute renal failureassociated with the histological changes now re-ferred to as diffuse D.T.N. These are a shut-down of the renal circulation, obstruction oftubules, and an unselective reabsorpt.on of theglomerular filtrate (Lucke, 1946; Bull et al., 1950).In addition leakage of glomerular filtrate throughdamaged tubular walls, producing interstitialoedema and a rise of intrarenal pressure, has beenpostulated to explain the persistence of theoliguria (Oliver et al., 1951 ; Oliver, 1953 ; Bulland -Dible, 1953). This is unlikely for two relatedreasons. First, the renal capsules were under ten-sion in both the oliguric and non-oliguric casesof acute uraemia associated with diffuse D.T.N.,and, secondly, no difference in the histologicaldegree or extent of the interstitial inflammatorychanges including interstitial oedema could bemade out between the two clinical forms.Haemoglobinuria.-Among the many nephro-

toxic substances which have been incriminated areheme compounds. Intravascular haemolysis com-monly occurs after severe burning, and if theplasma haemoglobin level rises above a thresholdvalue (120 to 140 mg. %) haemoglobinuria results.This may be slight and easily overlooked or deepred or reddish-brown and readily recognized; itmay be transient, or it may last one or two daysor longer. Thus the recording of clinically ob-served haemoglobinuria is an index of its severity

and the incidence of known haemoglobinuria willreflect the frequency of severe haemoglobinaemia.

In Table IV the frequency of haemoglobinuriaamong patients with and without tubular necrosisis set out. On the one hand 12 out of 16 patients(75%) with diffuse D.T.N. had haemoglobinuria:13 of the total number were known to have de-veloped renal failure (oliguric or non-oliguric),and in 11 of them (84%) haemoglobinuria wasknown. On the other hand only three out of 24patients (12.5%) without tubular necrosis hadhaemoglobinuria. The frequency of haemoglobin-uria among patients with focal D.T.N. (six outof 18 or 33%) and early P.T.N. (six out of 13 or46%) was intermediate. The overall incidence ofhaemoglobinuria in patients with renal failure was17 out of 24 (possibly 18 out of 26) or about 70%.Thus haemoglobinuria occurred in most of thepatients with acute renal failure, particularlyamong those who developed diffuse D.T.N. Con-versely haemoglobinuria was uncommon both inpatients without tubular necrosis and in those whodeveloped focal D.T.N., the great majority ofwhom did not develop renal failure.The manner by which haemoglobinuria induces

or predisposes to diffuse D.T.N. is not known withcertainty, but the histological appearance of manyhaemoglobin casts which focally block many distaltubules, render them focally necrotic, and inducetubular rupture cannot be disregarded as a patho-genetic factor. Since haemoglobin itself is said tobe non-toxic, other factors must also be requiredbefore a diffuse D.T.N. occurs. Oligaemia pro-ducing renal vasoconstriction and ischaemia maybe responsible and is under investigation.

SUMMARYA histological analysis of the kidneys of 86

burned patients showed a proximal tubularnecrosis in 17 and a distal tubular necrosis in 34subjects. The combined frequency of both kindsof tubular necrosis is similar in children, adults,and elderly sub,ects.

Proximal tubular necrosis occurs mainly inelderly subjects who had nephrosclerosis and iscommonly associated with severe oliguria. It ischaracterized by a widespread necrosis of theproximal convoluted tubules, and this was prob-ably the result of vasoconstriction of the main renalarteries or of a disturbance of the intrarenal bloodflow.

Distal tubular necrosis mainly affects childrenand younger adults and may or may not be asso-ciated with acute renal failure. If the condition

293

copyright. on S



j.com/

J Clin P


ber 1956. Dow

nloaded from

http://jcp.bmj.com/

294 S. SEVITT

affects many nephrons it is termed diffuse D.T.N.and this corresponds to Lucke's lower nephronnephrosis.

Diffuse D.T.N. was found in 16 patients andwas associated with anuria or severe oliguria oralternatively with a non-oliguric form of acuteuraemia. No histological differences could befound between the oliguric and non-oliguric forms.Haemoglobinuria was usual and may be of aetio-logical significance.Focal D.T.N. which affected few nephrons was

found in 18 patients. It usually affected children,was infrequently associated with acute renalfailure, and was rarely of clinical importance.

I am indebted to Dr. de Wardener for sending mehis sections of dogs' kidneys, to various colleagues fortheir interest, to my secretary, Mrs. M. Swinden, andto Mr. D. Gibb, A.I.M.L.T., who prepared many ofthe sections and helped with the photomicrography.The work was carried out in my capacity as a part-time member of the scientific staff of the MedicalResearch Council.

REFERENCESBarac, G. (1946). C. R. Soc. Biol. (Paris), 140, 580.-(1948). Ibid., 142, 547.Bratton, A. B. (1941). Lancet, 1, 345.Brown, C. E., and Crane, G. L. (1943). J. Amer. med. Ass., 122, 871.Bull,G. M.,andDible,J. H. (1953). In Recent Advances in Pathology,

6th ed., ed. G. Hadfield, p. 284. Churchill, London.- Jackes, A. M., and Lowe, K. G. (1950). Clin. Sci., 9, 379.Bywaters, E. G. L., and Beall, D. (1941). Brit. med. J., 1, 427.- and Dible, J. H. (1942). J. Path. Bact., 54. 111.Dunn. J. Shaw, Gillespie, M., and Niven, J. S. F. (1941). Lancet, 2,

549.- and Montgomery, G. L. (1941). J. Path. Bact., 52, 1.Erb, I. H., Morgan, E. M., and Farmer, A. W. (1943). Ann. Surg.,

117, 234.Gibson, T. (1944). Spec. Rep. Ser. med. Res. Coun. (Lond.), No. 249,

pt. vi, p. 192.Goodpastor, W. E., Levenson, S. M., Tagnon, H. J., Lund, C. C.,

and Taylor, F. H. L. (1946). Surg. Gynec. Obstet., 82, 652.Luck6, B. (1946). Milit. Surg., 99, 371.Monsaingeon, A., Tanret, P., and Daussy, M. (1949). Presse mdd.,

57, 1221.Oliver, J. (1953). Amer. J. Med., 15, 535.-MacDowell, M., and Tracy, A. (1951). J. clin. Invest., 30, 1305.Pack, G. T. (1926). Arch. Path. (Chicago), 1, 767.Page, I. H. (1943). Amer. J. Physiol., 139, 386.Sevitt, S. (1956). Journal of Clinical Pathology, 9, 12.Shen, S. C., Ham, T. H., and Fleming, E. M. (1943). New Engl. J.

Mfed., 229, 701.Smith, H. W. (1951). The Kidney, p. 778. Oxford University Press,

New York.Teschan, P. E., Post, R. S., Smith, L. H., Abernathy, R. S., Davis,

J. H., Gray, D. M., Howard, J. M., Johnson, K. E., Klopp, E.Mundy, R. L., O'Meara, M. P., and Rush, B. F. (1955). Amer.J. Med., 18, 172.

Trueta, J., Barclay, A. E., Daniel, P. M., Franklin, K. J., andPrichard, M. M. L. (1947). Studies of the Renal Circulation.Blackwell, Oxford.

Wardener, H. E. de (1955). Lancet, 1, 580.Wertheim (1867). Wien. med. Pr., 8, 1237 (quoted by Pack, G. T.). copyright.

on Septem

ber 4, 2020 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.9.4.279 on 1 N

ovember 1956. D

ownloaded from

http://jcp.bmj.com/

tubular and tubular necrosis in thekidneys ofburned … · "acute tubular necrosis." the...

Documents