Twenty women with a history of recurrent UTI, defined as at least three UTIs in the past year (subjects with two UTI’s in the past six months but not three in the past year were also enrolled) provided informed consent at the outpatient clinic of the University of Wisconsin Hospital and Clinics, and were enrolled in the trial approved by the University of Wisconsin Health Sciences Institutional Review Board. The number of subjects was based on achieving a statistical power of at least 80% for each of two parameters, infection-free status at the end of the study, and changes in the E. coli strain diversity measure. A six-month infection-free incidence of 60% in 20 UTI-susceptible women consuming SDC compared to a 20% incidence in untreated women would provide 80% power with p = 0.05 using a two-tailed t-test. Similarly, analysis of the diversity of E. coli strains isolated from 20 subjects pre- and post-consumption would have at least 80% power to detect a 0.66 standard deviation shift in the Kullback Liebler (KL) diversity measure  using a two-tailed t-test conducted at a significance level of 0.05. The mean age of participants was 37, with a range of 18 to 64, and the mean number of UTIs in the past six months was 2.4 with a range of two to five. Exclusion criteria included immune-compromising diseases, co-morbidities, chronic inflammatory bowel disease, diabetes, or cranberry allergy. Although target enrollment for the study was 20 subjects, data from 3 subjects were removed from the analysis due to one patient with exclusionary criteria realized after enrollment, one patient with sample contamination, and one patient with UTI contracted within one week of study enrollment.
A two-week antibiotic washout period of non-prophylactic antibiotics (i.e. full treatment course for acute UTI) was required before beginning the study. Women who were taking daily prophylactic antibiotics, however, were allowed to participate in the study, provided they met the inclusion criteria. A total of three women who were taking prophylactic antibiotics were enrolled in this study. The women in the study consumed one serving (42 g) of SDC, provided by Ocean Spray Cranberries Inc., each day for two weeks. Rectal swabs were collected at the time of study entry to sample intestinal bacterial flora, and SDC consumption began the following day. Rectal swabs were again collected after two weeks of SDC consumption, and within one day of final SDC consumption.
Clinical efficacy was determined by 1.) Comparison of six-month UTI rates pre- and post-consumption, and 2.) Kaplan-Meier analysis of the time until a patient’s first UTI since beginning the study compared to a previous control group with similar inclusion/exclusion criteria. The six-month UTI rate was based on patient self-reports and documentation of the total number of UTIs in the year prior to the study and six-months since beginning the study. Patients also reported the date of their first UTI since the start of the study.
The presence of E. coli, Enterococcus sp., Klebsiella sp., S. Saprophyticus, S. aureus, and Proteus sp. were quantified as colony forming units (CFU) per ml from rectal swab eluates on BBL™ CHROMagar Orientation media (24hr growth at 37°C). Lactobacillus was quantified as CFU/ml on deMan Rogosa Sharpe (MRS) Lactobacillus agar and grown anaerobically for 72hr at 37°C. In addition, rectal swab eluates were streaked separately onto eosin methylene blue agar to isolate E. coli, and the last five colonies at the end of the streak area, including any other morphologically distinct colonies, were stored for further analysis. This method provides ≥97% probability of capturing the predominant strain in the sample .
E. coli strain heterogeneity was determined by Diversilab™ DNA fingerprinting based on ERIC repetitive PCR . Five E. coli strains were isolated from rectal swabs pre- and post-consumption, and the number of unique fingerprints was determined. The KL diversity measure was used to determine changes in intestinal E. coli heterogeneity pre- and post- SDC consumption .
All isolates were confirmed as E. coli by amplification of the glutamate decarboxylase (gad) gene in addition to the colorimetric reaction on BBL™ CHROMagar Orientation media . Mulitplex PCR was used to determine phylotype (A, B1, B2, or D) and the presence of six pathogenicity islands (PAIs) and 15 VFs related to iron-uptake, adhesion and toxicity [20–22]. The PAIs tested were from E. coli J96 (IIJ96), E. coli CFT073 (ICFT, and IICFT) and E. coli 536 (I536, II536, and IV536) . The individual siderophore genes tested included fyuA, iroN
, and iutA . The adhesion genes tested were sfa, papG allele II/III, sfaS, and iha. The toxins included hlyA, cnf1, pic, tsh, sat, and tosA, and other genes tested were K1 and iss.
Two separate statistical analyses were used to evaluate clinical efficacy. A two-tailed paired t-test and mean differences with standard errors (SE) was used to determine significant differences in the six-month UTI rates pre- and post-consumption. A log-rank test was used to compare the time until first UTI between the current study group and two placebo groups from previous clinical trials. The two placebo groups from the previous trials were pooled (p = 0.83).
Significant differences in the numbers (CFU/ml) of bacteria present pre- and post-consumption were determined using the Wilcoxon matched pairs signed rank test using log-transformed values to calculate geometric means with 95% confidence intervals (CI), and the presence of VFs was expressed as percentages.