Published online July 2, 2007
PEDIATRICS Vol. 120 No. 1 July 2007, pp. 249-250 (doi:10.1542/peds.2007-1233)

This Article Extract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Request Permissions Citing Articles Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Garin, E. H. Articles by Young, L. Search for Related Content PubMed Articles by Garin, E. H. Articles by Young, L. Related Collections Genitourinary Tract Social Bookmarking                         What's this?

LETTER TO THE EDITOR

Much Pain, Little Gain From Voiding Cystourethrogram After Urinary Tract Infection

Eduardo Humberto Garin, MD
Department of Pediatrics

Linda Young, PhD
Department of Biostatistics
University of Florida
Gainesville FL 32610

To the Editor.—

We have read the thoughtful comments on our article¹ on the significance of primary vesicoureteral reflux and urinary antibiotic prophylaxis after acute pyelonephritis. Here we want to address the issues of methodologic flaws referred to by Wald.²

AN INTENTION-TO-TREAT ANALYSIS WAS NOT PERFORMED, WEAKENING THE RESULTS OF THE STUDY

To address this concern, we have filled in missing values for each group assuming a worst-case scenario. That is, patients in the nonprophylaxis group who did not complete the study were all assumed to have presented a recurrence of the urinary tract infection and renal scars; patients in the prophylaxis groups who did not complete the study had no urinary tract infection or renal scars. With respect to the recurrence of urinary tract infections (any kind, because we could not assume the localization of the infection), 3 primary questions of interest were considered. First, for patients with vesicoureteral reflux, was there a significant difference in the proportions with recurring infection or renal scars and those with or without prophylaxis? The new analysis showed that there was no significant difference for either infection (P = .67) or renal scars (P = .73). Second, for patients who did not receive prophylaxis, was there a significant difference in the proportions of those with and without vesicoureteral reflux who experienced infection or renal scars? The reanalysis of the data showed that there was no significant difference in infection (P = .83) or renal scars (P > .99). Third, was there a significant difference in the proportion of patients who became infected in the vesicoureteral-reflux-with-prophylaxis and the no-vesicoureteral-reflux-without-prophylaxis groups? Again, applying the new analysis of the data, there was no significant difference in infection (P = .83) or renal scars (P = .73). With this most severe method of analysis, there is still strong support for our primary conclusions.

ANTIBIOTIC TREATMENT OF THE ACUTE EPISODE WAS NOT STANDARDIZED

All patients were initially treated with intravenous antibiotics according to sensitivity. The duration of intravenous therapy varied from center to center. However, the minimum number of days on intravenous therapy was similar to that used in the Hoberman et al study.³ All patients received a 14-day course of antibiotics in the case of acute pyelonephritis. Finally, our renal-scar rate is quite similar to that observed by Hoberman et al.

PLACEBO WAS NOT ADMINISTERED TO THE CONTROL GROUP

Given the end points to our studies, we fail to see that administering placebo to the control group would have influenced the rate of recurrence of infection or formation of renal scarring. In addition, to assess for compliance with medication, only determination of urinary antibiotic concentration is considered acceptable, as other studies have clearly shown. It is not enough to have the placebo available. It is important to note that in the current study proposed by the National Institutes of Health and quoted by Wald,² although a placebo is administered, no measurements of urinary antibiotics are performed.

NEITHER SUBJECTS NOR PHYSICIANS WERE BLINDED TO THE TREATMENT ASSIGNMENTS

We understand that blinding is necessary in certain types of studies in which the Hawthorne effect may play a role. By definition, the Hawthorne effect refers to the improvement seen in patients just because of the special attention paid to the patients during their follow-up. It refers mostly to subjective information. In our opinion, the absence of blinding does not introduce a bias, because the end points in this study (documentation of urinary tract infection and renal scars) were beyond the control of the patients. In addition, the patients who received prophylaxis did not do better than those who did not receive it.

We completely agree that our results should apply only to patients with mild-to-moderate vesicoureteral reflux. We also agree that complete validation of our results will come only when other investigators confirm them.

REFERENCES

1. Garin EH, Olavarria F, Garcia Nieto V, Valenciano B, Campos A, Young L. Clinical significance of primary vesicoureteral reflux and urinary antibiotic prophylaxis after acute pyelonephritis: a multicenter, randomized, controlled study. Pediatrics. 2006;117 :626 –632[Abstract/Free Full Text]
2. Wald ER. Much pain, little gain from voiding cystourethrograms after urinary tract infection: in reply. Pediatrics. 2006;118 :2251 –2252[Free Full Text]
3. Hoberman A, Wald ER, Hickey RW, et al. Oral versus initial intravenous therapy for urinary tract infection in young febrile children. Pediatrics. 1999;104 :79 –86[Abstract/Free Full Text]

---

PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics

CiteULike    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This Article Extract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Request Permissions Citing Articles Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Garin, E. H. Articles by Young, L. Search for Related Content PubMed Articles by Garin, E. H. Articles by Young, L. Related Collections Genitourinary Tract Social Bookmarking                         What's this?