The Swedish Reflux Trial in Children: IV. Renal Damage

Results 

Abnormal DMSA findings at entry were seen in 124 children (61%), of whom 18 (15%) had bilateral uptake defects. Generalized renal damage was found in 30 of 128 girls (23%) and in 44 of 75 boys (59%) (p <0.0001). Two-year DMSA scan was done in all except 2 of the 203 children. Hospital fear was the reason for the omitted investigation in the 2 children, of whom 1 had a class 1 uptake defect at entry and 1 had normal kidneys.

Renal status deterioration, that is new damage in previously unscarred kidney areas and greater than 3% decreased relative function in a kidney with uptake defects at entry during the 2-year period, was observed in 17 girls and 7 boys, including 4 of 68 (6%) on prophylaxis, 8 of 65 (12%) with endoscopic therapy and 12 of 68 (18%) on surveillance. These differences were not statistically significant (p = 0.11).

Deterioration was seen in 15 of 49 children (31%) with and in 9 of 152 (6%) without recurrent febrile UTIs (p <0.0001, table 1). Median time from the last reported febrile UTI to followup scan was 225 days. In 4 cases it was less than 6 months (79, 81, 83 and 114 days, respectively). Three cases showed a distinct new uptake defect with retraction of the renal contour, which is typical for a permanent scar, while 1 girl on surveillance with 4 febrile UTIs had a more diffuse uptake defect at followup. There was no difference in renal deterioration between children with grade III or IV VUR at study entry (each 12%).

Table 1. Febrile recurrences and renal status deterioration on 2-year DMSA scintigraphy

		No. Recurrences⁎
		0	1	2	3	4	Totals
	No. deterioration:
	No	143	22	10	2	0	177
	Yes	9	9	3	2	1	24
	Totals	152	31	13	4	1	201
	No. new renal damage:
	No	148	25	10	3	0	186
	Yes	4	6	3	1	1	15
	Totals	152	31	13	4	1	201

⁎ Followup DMSA scan not done in 2 patients (p <0.0001).

We further analyzed the 13 girls and 2 boys with new renal damage (fig. 1). Eight of the 13 girls were on surveillance, 5 had endoscopic treatment and none were on prophylaxis (p = 0.0155). New scarring occurred significantly more often in the surveillance than in the prophylaxis group (p = 0.0054). However, in 2 girls with febrile recurrences during the study period a febrile UTI developed between the first DMSA scan and randomization. Thus, new damage before randomization could not be ruled out in these cases. After excluding those patients there were still significantly more girls with new scars on surveillance than on prophylaxis (p = 0.0258). The difference between the endoscopic and prophylaxis groups did not attain significance and there was no difference between the endoscopic and surveillance groups (p = 0.0551 and 0.5477, respectively). New damage was seen in 11 of 49 children (22%) with vs 4 of 152 (3%) without recurrent febrile UTIs (p <0.0001, table 1). New damage developed in 9 children with normal kidneys and in 6 with abnormal kidneys at study entry.

View Large Image Download to PowerPoint

Figure 1. New renal damage in children with dilating VUR by treatment arm.

Analysis of individual kidneys showed that deterioration (progress of initial uptake defects or development of new scars) occurred only in kidneys drained by ureters with grade III or IV VUR (table 2). In patients with bilateral VUR deterioration always occurred in the kidney with the most severe VUR except in 1 patient (fig. 2). In this girl with normal kidneys, and grade III and II VUR at study entry 2 febrile UTIs and bilateral scars developed.

Table 2. VUR at study entry

	VUR Grade	No. Kidneys⁎	No. Progression	No. New Damage
	NoVUR	92	0	0
	I	11	0	0
	II	44	0	1
	III	173	7	10
	IV	83	6	5
	Totals	402	13	16

⁎ Followup DMSA scan not done in 2 patients.

View Large Image Download to PowerPoint

Figure 2. DMSA scan at study entry (A) and at 2-year followup (B) in girl with bilateral renal damage

Discussion 

In this RCT the rate of new renal damage in boys was low. New damage developed in 1 boy with endoscopic treatment and 1 on surveillance. In girls the rate was higher (13 of 126 or 10%) and there was a significant difference between treatment groups with new damage most common in controls on surveillance. New damage was observed in 19% of girls on surveillance, in 12% with endoscopic treatment and in none on prophylaxis.

The febrile UTI rate differed significantly between treatment groups in girls, including a recurrence rate of 57% in the surveillance, 23% in the endoscopic and 19% in the prophylactic groups.17 There was a strong association between recurrent febrile UTIs and new renal damage in girls but recurrent UTIs and new damage were unusual in boys. It is well-known that UTIs are less common in boys after age 1 year.18, 19

In recent years 3 RCTs have compared prophylaxis vs no prophylaxis in regard to the development of kidney damage.20, 21, 22 Garin et al investigated 218 children with no or grade I–III VUR, including 37 with grade III.20 New damage was detected in 13 patients (6%) and there was no difference between those who did and did not receive prophylaxis. However, new damage developed in 5 of 37 patients with grade III VUR vs 8 of 173 with no or grade I–II VUR (Fisher's exact test p = 0.049). Pennesi et al investigated 100 children with grade II–IV VUR and found new scarring in none.21 Ten patients had worsening on DMSA scan between study start and followup without a difference between the 2 groups. All 10 patients had grade IV VUR. Montini et al studied 338 children with no or grade I–III VUR and found new scars in 4 of 295 (1.4%) with no difference between the prophylaxis and control groups.22 However, only 40 children had grade III VUR, including 14 controls and 26 on prophylaxis. No conclusions could be drawn on the efficacy of prophylaxis in children with higher VUR grades. These RCTs indicate that in children with no or nondilating VUR the risk of renal damage is low and antibiotic prophylaxis has no value.

In the Swedish trial we only included children with grade III (126) and IV (77) VUR, and found significantly more girls with new renal damage in the surveillance group. To our knowledge our series is the largest RCT to date in children with dilating VUR. Our results must be validated by other RCTs, such as the Randomized Intervention for Children With Vesicoureteral Reflux study, in which 600 children with grade I–IV VUR are being allocated to prophylaxis or placebo.23

Most analyses in our study were done on a patient basis. However, we also assessed each renal unit separately regarding the relation between VUR grade and kidney status at followup. An interesting finding was that new damage as well as the progression of previously observed renal uptake defects was only seen in kidneys drained by ureters with dilating VUR with 1 exception. There was no difference in this respect whether VUR was grade III or IV at study entry. Previously unscarred kidneys were as vulnerable to new renal damage as those with uptake defects at entry.

Renal damage progression may be a consequence of new insults to the kidney or a long-term process in a damaged kidney that does not reflect events that occur during the study period. Compensatory growth of the contralateral kidney may also be misinterpreted as progression since the difference in side distribution increases. Thus, we focused on new scarring as a more reliable end point.

General parenchymal damage at entry was seen in 59% of boys and in 23% of girls, supporting the fact that congenital damage is predominantly seen in boys. New damage was mainly seen in girls (13 girls vs 2 boys), supporting the concept that acquired renal damage is more prevalent in girls.8

A study limitation was the lack of a defined time limit between the first DMSA scan and randomization. The median interval was 49 days but in a few patients it was long with the risk of new UTIs, as in 2 girls with new renal damage which, thus, may have developed before randomization. Another limitation was the lack of placebo treatment in controls. Study strengths include a larger patient population than previously described, age and VUR grade homogeneity, close monitoring during followup to ensure adequate diagnosis of all febrile UTIs, good adherence to protocol with DMSA scan repeated in 99% of children and intent to treat analysis.

Conclusions 

In boys the rate of new renal damage was low. It was significantly higher in girls and most common in controls on surveillance. There was also a strong association between recurrent febrile UTIs and the development of new renal damage in girls.