Acute renal failure in paediatric population in the tropics.
The syndrome of acute renal failure (ARF) in children results from many causes. The prognosis of any case depends on the cause and the speed with which the condition is detected and treated. We wish to record our experience in 51 paediatric cases of ARF seen at the K.E.M. Hospital, Bombay during a 3 year period from 1981 to 1983. According to the current hospital practice, the paediatric group is limited to children who have not completed 12 years of age. The diagnosis of ARF was based on rapidly progressing azotemia (rise of serum creatinine by at least 0.5 mg/dl/day and blood urea nitrogen by 10 mg/dl/day usually but not always associated with oliguria.
[Table - 1] shows the age and sex distribution of 51 patients. The youngest child was 6 months old and the oldest, 12 years, the average being 63.3 months. Each patient underwent a detailed clinical, hematological and biochemical examination. Urine culture was done in all patients in whom urine could be obtained. In patients with bleeding tendency, a detailed coagulation study was done. The kidney size was assessed by K.U.B., I.V.P. or ultrasonography. Histopathological examination of the kidney was done on biopsy or autopsy material in some cases.
The etiological factors responsible for ARF in our study are listed in [Table - 2].
The most common presenting feature was oligo-anuria (94.1%). Three patients (5.9%) presented with non-oliguric renal failure. Bleeding tendency was seen in 14 patients (27.5%), the commonest site being the G.I. tract. The neurological manifestations observed were altered sensorium (60.8%), convulsions (29.4%) and papilloedema (15.7%). The other presenting features were hypovolemia, water-logging, hypertension, congestive failure and acidosis.
[Table - 1] also shows survival in different age groups. In general, the survival was poor in those aged less than three years as compared to those aged more than 3 years. The survival was better in patients with a shorter duration of oligo-anuria. All the three patients who presented with non-oliguric renal failure survived. Similarly, there was definite relation between the delay in seeking specialized treatment and survival [Table - 3] When the delay was short, the survival was better.
Out of 15 patients who had convulsions, only 5 survived (33.3%). Eight patients presented with papilloedema while 3 developed the same during their stay in the ward. Only one patient survived out of these total 11 cases. The survival was better in cases of acute glomerulo-nephritis (71.4%), while in cases of gastro-enteritis and hemolytic uremic syndrome, the survival was 33.3%.
Blood urea nitrogen, plasma creatinine, bicarbonate and potassium were estimated in our patients at the time of their admission. These values were later correlated with the mortality at the end of the study. It was noted that azotemia, acidosis and hyperkalemia were more in patients who expired as against those who survived [Table - 4] The differences were, however, not statistically significant.
Histopathological study could be done in 33 cases-mostly in autopsy cases. Tubular necrosis was seen in 22, cortical necrosis in 6, fibrin thrombi in the glomeruli and the arterioles in 4 and glomerular lesions in 5. In 2 cases, there was a combined picture of fibrin thrombi, patchy cortical necrosis and tubular necrosis. Amongst the 5 cases of glomerular lesions, 2 had proliferative and exudative glomerulo-nephritis whereas one each had crescentic glomerulo-nephritis, focal and segmental proliferative glomerulo-nephritis (in a case of Henoch-Schonlein purpura) and diffuse proliferative glomerulonephritis going into sclerosis (in a case of infective endocarditis).
Three children with non-oliguric renal failure were managed conservatively, 44 of the oliguric patients were given peritoneal dialysis while 4 were given haemodialysis. In 8 children where peritoneal dialysis was not effective, hemodialysis was resorted to subsequently. The mortality in the three years under consideration was /11 in 1981, /18 in 1982 and /22 in 1983.
Acute gastro-enteritis, acute glomerulonephritis and hemolytic-uremic syndrome were the most common causes of ARF in our study. Choudhry et al also reported these causes as common etiological factors. On the other hand, in U.S.A. and Europe, the important causes include complications of major surgery, trauma and hemolytic uremic syndrome.,,
Acute renal failure in children is associated with oliguria in 90% of cases. However, about 10% of patients excrete urine volumes upto 1 litre/m/day or more. In our study also, only 3 patients (5.9%) presented with non-oliguric renal failure. Bleeding tendency and neurological manifestations were noted as in other studies.,, The survival was better in older children than in younger ones. This is probably because primary glomerular disease is more common in older children than in younger ones in whom renal failure is usually secondary to some other underlying precipitating factor (commonly gastro-enteritis). We have found survival to be better with primary glomerular disease. Even Hodson et al reported better prognosis with primary nephrologic disorders. Moreover, technical problems of dialysis are less in older children than in younger ones, making dialysis more effective in older children.
Prolonged duration of oliguria and total anuria have ominous prognostic significance. We observed that survival was better in those with a shorter duration of oligo-anuria. All 3 patients who presented with non-oliguric renal failure survived. In patients who presented with glomerulo nephritis, the outcome was found to be better despite prolonged oligo-anuria, explaining higher percentage of survival in the group with prolonged oligo-anuria [Table - 3].
Early and frequent dialyses lead to improved survival in patients of acute renal failure. In our study, survival was better when the delay in seeking specialized treatment was shorter. Hence, early referral to a specialized centre is advocated to ensure best results. There were 5 patients where the delay was more than 7 days and 3 out of these survived [Table - 3]. One was a case of post-streptococcal glomerulo-nephritis while other two were cases of acute tubular necrosis, both of whom were clinically stable, had no associated complications and started passing urine within a day of admission. Out of the 2 patients who expired, one was a case of acute tubular necrosis with severe chest infection and the other was a case of hemolytic uremic syndrome with bleeding tendency and neurological complications which we have found to be associated with poor prognosis. The neurological manifestations of acute uremia are poorly understood. Altered sensorium, convulsions and papilloedema were the features that we commonly observed and we found them to be of ominous prognostic significance. Attempts at isolation of single, generally accepted, neurotoxic factor in uremia have been unsuccessful. However, it is evident that urea is not responsible as shown by the improvement which follows dialysis of patients against a bath of high urea concentration.
Hodson et al observed that of the 30 patients with primary nephrologic disorder, 27 survived whereas only 4 out of 23 patients with acute tubular necrosis survived. These authors stated that it is not the renal failure per se but the under lying disorder which in large measure determines the outcome. In our study, we found survival in patients of acute nephritis to be 71.4% as compared to 33.3% in patients with acute gastro-enteritis and hemolytic uremic syndrome.
The difference between patients who survived and those who died regarding azotemia, acidosis and hyperkalaemia on admission was not statistically significant. Further, with the advent of dialysis, the final outcome in cases of acute renal failure depends more on the underlying illness and secondary complications (infection, neurological and hematological complications) rather than on factors such as azotemia, acidosis or hyperkalaemia.
The most common histopathological finding was tubular necrosis. This was mostly seen with acute gastro-enteritis which was the commonest cause of ARK Fibrin thrombi with patchy cortical necrosis was the picture seen in cases of hemolytic uremic syndrome.
Peritoneal dialysis is the preferred form of therapy in children., The most important factor limiting the use of hemodialysis in paediatric patients is the difficulty in obtaining vascular access. Secondly, for achieving efficient dialysis, a blood flow of 200-300 ml/min through the shunt is regarded to be optimum. To achieve this in children, proximal vessels will have to be cannulated which would produce greater increase in cardiac output with deleterious effects on cardiac function. Thirdly, magnitude of fluctuation in blood volume is greater in younger children, making hemodialysis more hazardous. In our study, out of 48 patients, who were dialysed, 44 were given peritoneal dialysis while only 4 were given hemodialysis and they were in the older age group.
The overall mortality in the present series was 60.8%. This is higher than that reported by other workers. This is partly because of a series of 11 cases that we had in 1982. In that year, we saw these children coming from a common locality over a period of 4 months, ten of them being less than 2 years of age. Three patients had no precipitating factors while in the remaining 8, a short history of fever, vomiting or loose motions could be obtained. All of them presented with total anuria. Six of these 11 patients developed convulsions and papilloedema. They were given peritoneal dialysis, but after an initial improvement all of them expired. Despite extensive investigations including virological study, the etiology could not be determined in any. Histopathology of autopsy specimens revealed tubular necrosis in the kidney and necrosis of hepatocytes around the central vein in the liver. Mehta et al reported a similar clinical presentation around the same time in adults in and around Jaipur in Rajasthan. Histopathology in those cases revealed tubular necrosis in the kidney. Mortality was 80% in their study. It is possible that an unidentified toxin was responsible for renal and hepatic damage.
In conclusion, it would be worthwhile emphasising that even today, acute gastroenteritis is a major cause of ARF in tropical countries like ours. If this condition, is treated early, the occurrence of ARF can be prevented. Hence, it is extremely important that this condition should be treated in its early phase with adequate fluid replacement and control of the primary medical problem. If the patient develops ARF, an early referral to a specialised centre with dialysis facilities, can significantly improve the final outcome.