Archives of Disease in Childhood, 1986, 61, 1155-1158

Investigation of

Urinary tract infection (UTI) is common in infants Kingdom recommend a slightly less invasive and children, with a female preponderance except approach, in which all children are investigated by during early infancy. In an early report from urography (or, in appropriately equipped centres, University College Hospital, London, Smellie et al abdominal ultrasonography supplemented when described 200 consecutively diagnosed children with necessary by a renal radionuclide scan), micturating UTI, of whom 188 were investigated by excretory cystourethrography being reserved for children urography or micturating (voiding) cystourethro- under 5 and patients in whom the urogram is graphy, or both.' Abnormalities were found in 50% abnormal or who suffer from frequent in- and in 35% of those investigated after their first fections. 11 12 infection (47% of boys and 28% of girls). In When experts disagree, how is the non-specialist particular, chronic atrophic pyelonephritis (also to decide how to investigate the rather small number known as reflux nephropathy) was detected in 13% of children presenting to him with urinary infection of the total; nearly all of these were children with (two a year for the average general practitioner'3)? bacteriologically proven or clinically suspected mul- To answer this question, it is necessary to under- tiple infections. Vesicoureteric reflux was present in stand the capabilities and, perhaps more import- 30% of investigated subjects but in 85% of those antly, the limitations of the various techniques with pyelonephritis. available for imaging the urinary tract, so that the Numerous subsequent studies have yielded results most appropriate test or combination of tests can be in broad agreement with these, the proportion of selected to suit the needs of each particular case. investigated children showing some radiological abnormality ranging from 25 to 55% and the Excretory urography proportion with reflux from 8 to 52%.2-9 In at least some of these reports a significant proportion of Urography is a powerful diagnostic tool that is patients had severe (grade III or IV) reflux in the capable of providing useful information about the presence of a normal urogram, suggesting that entire urinary tract except the . The early excretory urography alone is not adequate to (nephrogram) phase shows the renal outlines accur- exclude serious and potentially damaging reflux.2 69 ately, particularly when enhanced by tomography, Since reflux is known to be causally associated with and the slightly later phase delineates the pyelonephritis, which itself is responsible for roughly pelvicalyceal (collecting) systems. The coarse, 20% of the end stage renal failure occurring in irregular scarring of pyelonephritis is readily recog- European patients below 40 years of age,'( the early nisable, as is the clubbed calyx and overlying cortical identification of patients with reflux might offer the thinning of the more localised lesion. Radio-opaque prospect of preventing some cases of renal failure by stones will be seen on the control ('scout') film taken appropriate intervention before severe damage has before contrast is injected and radiolucent ones occurred. In addition, some children with UTI have show as filling defects. Obstructive lesions such as surgically correctable abnormalities of the urinary obstruction of the pelviureteric junction are well tract other than reflux, such as obstructive lesions or shown. In the non-obstructed system renal function urolithiasis. can be roughly inferred from the amount of opaci- On the basis of arguments such as these, many fied tissue seen on the nephrogram, in the absence paediatricians, radiologists, and urologists teach of diffuse parenchymal disease such as glomeru- that all children in whom urinary tract infection has lonephritis. A reasonably good view of the been diagnosed should be studied by urography and and bladder (full and empty) can normally be micturating cystourethrography, the former inves- obtained. tigation being necessary to obtain a reliable image of Advantages of the method include familiarity, the renal parenchyma, collecting systems, and lower universal availability, and relative ease of inter- urinary tract and the latter to identify or exclude pretation. No other technique approaches it for reflux;3 7 9 this view seems to be particularly widely showing the fine detail of the pelvicalyceal systems. held in North America. A few advocate It probably gives a better overall view of the renal in most or all children,2 84 but this is certainly a tract than any other single test. Disadvantages minority opinion. Some experts in the United include the risk of allergic reactions to contrast 1155 1156 Haycock material, which although usually minor can be 99mTc dimercaptosuccinic acid (DMSA) scan17 serious, the discomfort of the injection in small children, the entailed irradiation of the gonads, DMSA is selectively taken up by renal tubular especially in girls, and the dependence of the image epithelial cells; the emitted y radiation can be used quality on good renal function. Satisfactory pictures to form an 'autoradiograph' of the kidneys, which are difficult to obtain in infants and especially the shows the renal outlines fairly well, though lacking newborn. Small renal scars may be missed if located the definition of urography. on the anterior or posterior aspect of the . Focal scars shows as 'holes' in the image and will be seen even if they are on the anterior or posterior Micturating (voiding) cystourethrography aspect of the kidney, where they are likely to be missed by urography. By counting the radioactivity Micturating cystourethrography is the definitive over the two kidneys, differential renal function can method of showing reflux. It also gives a higher be estimated. The radiation dose is less than definition view of the bladder than excretory urog- urography; reactions to the radionuclide are not raphy, and in the lateral projection, during voiding, seen. The technique gives good results in neonates. the urethra is outlined in its full length. Its major disadvantage is the need for bladder catheterisation, 99mTc diethylenetriamine penta-acetic acid (DTPA) which is at best uncomfortable and at worst painful scan and frightening for small children. Infection may be introduced, and urethral strictures occassionally DTPA is handled by the kidney in the same manner occur. Nine serious complications of micturating as inulin and radiocontrast media such as sodium cystourethrography, including two fatalities, were iothalamate (Conray)-that is, it is filtered at the described in a series of 14 children reported by glomerulus and neither secreted nor reabsorbed by McAlister. 14 As with excretory urography, some the tubule. When labelled it therefore provides a gonadal irradiation is inevitable, especially in girls. 'radionuclide urogram' with the same phases as an It is by far the most unpopular procedure with excretory urogram: nephrogram followed by pyelo- patients and their parents of those considered here. gram with subsequent visualisation of ureters and bladder. Image definition is considerably inferior to Ultrasonography that obtainable by urography, and the technique should not be considered a substitute for it; by Given up to date equipment and an experienced measuring the transit time of radioactivity from operator, real time ultrasonography is capable of kidneys to bladder after an injection of furosemide, providing a great deal of information about the however, obstructive lesions such as obstruction of anatomy of the renal tract. The shape, size, and the pelviureteric junction can be detected. In a position of the kidneys can be determined with further refinement a secondary rise in activity over considerable accuracy; the degree of 'echogenicity' one or both kidneys coinciding either with bladder of the renal substance and the sharpness of corti- emptying or with a fall in bladder activity without comedullary differentiation provide a clue to the emptying may indicate the presence of reflux. presence of diffuse parenchymal disease or damage. Hydronephrosis, hydroureter, and ureterocoeles are Purpose of imaging studies reliably detected, as are bladder dilatation, trabe- culation, and other relatively gross abnormalities of Before submitting a patient to a series of imaging the lower urinary tract. Intermittent dilatation of the studies of the urinary tract, it is desirable to define retrovesical sometimes provides a clue to the as precisely as possible the questions being asked. presence of reflux, although normal ultrasonog- These will vary according to the age of the patient raphy certainly does not exclude it. Nor does it and the clinical circumstances of the case. They may exclude focal scarring of the kidneys due to include any or all of the following: pyelonephritis,'5 16 and of course it provides no (i) Is there any structural abnormality of the information at all about renal function. More than urinary tract-for example, obstruction, stone, any other test mentioned in this annotation, it is neuropathic bladder-predisposing to infection? 'driver dependent'; at the time of writing, relatively (ii) Are the kidneys scarred as a result of the few ultrasonographers have the skill, the experi- infection(s)? ence, or perhaps even the aptitude to obtain the (iii) Is reflux present and, if so, in what degree of maximum available information from renal ultraso- severity? nography, while the pictures themselves are difficult Questions (i) and (ii) can be answered with a high for the clinician to evaluate. degree of reliability either by excretory urography Investigation of urinary tract infection 1157 or by a combination of plain abdominal x ray, tomatic recurrences of infection were recorded'. ultrasonography, and DMSA scan. Question (iii) These cases were assembled from 23 centres over a can only be definitively answered by micturating span of 22 years. Therefore, new scars were seen in cystourethrography, and it is here that controversy only one asymptomatic patient per centre every 28 is centred: how important is it to detect every case of years-hardly a common event. reflux after the first infection? Put another way: is it possible to identify subgroups of children in whom Suggested scheme of investigation the risk of progressive renal damage after their first UTI is so low that cystourethrography can safely be The following proposed guidelines are similar to, omitted? I believe that with a proper understanding but not identical with, those recently put forward by of the epidemiology of chronic atrophic pyelone- Whitaker and Sherwood.23 Three age bands can be phritis, it can. identified. It is difficult to know the exact incidence of UTI in (1) Infants <1 year of age. All cases should be childhood, but a reasonable approximation may be investigated by micturating cystourethrography, made from a carefully conducted, prospective study ultrasonography, and DMSA scan. If apparatus for published by Dickinson in 1979.13 He found that or expertise in the last two are not locally available, 3*1/1000 girls and 1-7/1000 boys, aged 0-14, pre- excretory urography (intravenous urogram) can be sented annually with a first, symptomatic UTI; this substituted, but visualisation of the kidneys and corresponds to about 780 and 430 per million total ureters is often unsatisfactory. population, respectively. (An incidental, but very (2) Children aged 1-7 years. All children should important, observation made in the same study was be studied by either excretory urography or the that only 18% of children with dysuria and fre- combination of plain abdominal film, ultrasono- quency proved to have infected urine, a figure that graphy, and DMSA scan. agrees well with the 24% reported earlier by (3) Children aged 7 years and over. A plain film Williams;'8 this underlines the importance of and ultrasonography are sufficient. properly establishing the diagnosis at the first If excretory urography is selected as the initial presentation, the single most effective way of investigation a strong case can be made for delaying avoiding unnecessary investigations.) The incidence the study for six weeks after the infection has been of end stage renal failure due to pyelonephritis in diagnosed and treated. This is because the charac- Europeans aged <40 is in the region of 4-5 per teristic radiological appearances of pyelonephritis million. '( Combining these figures, the crude risk of are due to contraction of the scar tissue, which renal failure due to pyelonephritis in children replaces the damaged parenchyma, a process that presenting with a first, symptomatic UTI is not takes time; urography performed immediately after greater than 1% for boys and 0-5% for girls. This the infection may give a false negative result. In this must be an overestimate of the true risk as it case a small prophylactic dose of a suitable anti- assumes that detection and investigation of child- microbial agent should be continued after the hood UTI would have identified all those who would infection has been eradicated until the result of the later develop renal failure, which is not the case; urogram is known. many adult patients with end stage pyelonephritis A micturating cystourethrogram should be per- have no history of symptomatic UTI in childhood. formed in children of 1 year and over if: (i) the Furthermore, identification of those with reflux with infection recurs; (ii) the studies recommended or without pyelonephritis in childhood would not above are abnormal; (iii) the clinical features of the necessarily lead to prevention of renal failure in all first infection were suggestive of acute pyelonephri- cases. It is therefore likely that the probability that tis; or (iv) there is a known family history of chronic any individual child with a first UTI is at risk from atrophic pyelonephritis. If this scheme is followed, potentially preventable renal failure is of the order and provided that the recommended imaging studies of one in several hundred. are performed and interpreted by staff skilled in the The bulk of the published evidence suggests that investigation of children, most of those presenting scarring is present at the first investigation in most of with a first UTI after their first birthday can be those children in whom renal damage develops.'9 20 spared a micturating cystourethrogram without A recent study has shown unequivocally that new great risk of missing important reflux. scars can develop in both previously undamaged and damaged kidneys;2' as pointed out by Sherwood and References Whitaker,22 however, this is a rare event. In the Smellie JM, Hodson CJ, Edwards D, Normand ICS. Clinical study of Smellie et al new scars were seen in 74 and radiological features of urinary infection in childhood. Br children.21 In 56 of these, however, 'repeated symp- Med J 1964;ii:1222-6. 1158 Haycock

2 Wein AJ, Schoenberg HW. A review of 402 girls with recurrent 5 Kangarloo H, Gold RH, Fine RN, Diament MJ, Boechat MI. urinary tract infection. J Urol 1972:107:329-31. Urinary tract infection in infants and children evaluated by 3Saxena SR, Laurance BM, Shaw DG. The justification for early ultrasound. Radiology 1985;154:367-73. radiological investigations of urinary-tract infection in children. 16 Leonidas JC, McCauley RGK. Klauber GC, Fretzayas AM. Lancet 1975;ii:4034. Sonography as a substitute for excretion urography in children 4Hallett RJ, Pead L, Maskell R. Urinary infection in boys: a with urinary tract infection. AJR 1985;144:815-9. three year prospective study. Lancet 1976;ii: 1 107-10. 17 Merrick MV, Uttley WS, Wild SR. The detection of 5Cohen M. The first urinary tract infection in male children. Ain pyelonephritic scarring in children by radioisotope imaging. Br J J Dis Child 1976;130:810-3. Radiol 1980;53:544-56. 6 Shah KJ, Robins DG, White RHR. Renal scarring and Ix Williams CM. Urinary tract infection in children: general vesicoureteric reflux. Arch Dis Child 1978:53:21(-7. practice survey. Aust Fain Physician 1976;5:340-4. 7Middleton AW Jr, Nixon GW. The lack of correlation between 9 Hodson J. Reflux nephropathy: a personal historical review. upper tract changes on excretory urography and significant AJR 1981;137:451-62. . J Urol 1980;123:227-8. 21' Berg UB, Johansson SB. Age as a main determinant of renal McKerrow W, Davidson-Lamb N, Jones PF. Urinary tract functional damage in urinary tract infection. Arch Dis Child infection in children. Br Med J 1984;289:299-303. 1983;58:963-9. Blickman JG, Taylor GA, Lebowitz RL. Voiding cysto- 21 Smellie JM, Ransley PG, Normand ICS, Prescod N, urethrography: the initial radiographic study in children with Edwards D. Development of new renal scars: a collaborative urinary tract infection. Radiology 1985;156:659-62. study. Br Med J 1985;290:1957-60. Broyer M, Rizzoni G, Brunner FP, et al. Combined report on 22 Sherwood T, Whitaker RH. Initial screening of children with regular and transplantation of children in Europe, XIV, urinary tract infections: is plain film and ultraso- 1984. Proc Eur Dial Transplant Assoc 1984;5:55-79. nography enough? Br Med J 1984;288:827. Winterborn MH. The management of urinary tract infections in 23 Whitaker RH, Sherwood T. Another look at diagnostic path- children. Br J Hosp Med 1977; 17:453-61. ways in children with urinary tract infection. Br Med J 12 Smellie JM, Normand ICS. Urinary infections in children 1985. 1984:;288:839-41. Arch Dis Child 1985;61:895-90)5. 3 Dickinson JA. Incidence and outcome of symptomatic urinary GEORGE B HAYCOCK tract infection in children. Br Med J 1979Ji:133(-2. Evelina Department of Paediatrics, 14 McAlister WH, Cacciarelli A, Shackelford GD. Complications associated with in children. Radiology 1974; Guy's Hospital, 111:167-72. London SE] 9RT