Pilot Randomized Controlled Dosing Study of Cranberry Capsules for Reduction of Bacteriuria Plus Pyuria in Female Nursing Home Residents

To the Editor

Cranberry products represent a non-antimicrobial method for prevention of urinary tract infection (UTI). Cranberry proanthocyanidin (PAC), a type of condensed tannin, is the active ingredient in cranberry that inhibits adherence of P-fimbriated Escherichia coli (E.coli) to uroepithelial cells.^{1, 2} Previous cranberry studies for UTI prevention yielded conflicting results, likely because of variability of PAC dose and clinical populations studied.^{3, 4} In a clinical trial of 300ml of cranberry juice beverage daily (36mg PAC), older women (mean age 78.5 years) had a 58% reduction in the odds of having bacteriuria plus pyuria compared to controls.⁵ However, nursing home residents have difficulty ingesting the volume of juice that is necessary to prevent bacteriuria. Cranberry capsules are feasible to administer to nursing home residents, but their efficacy has not been demonstrated.⁶ In vitro, 36 to 108mg of PAC is efficacious at inhibiting bacterial adherence to uroepithelial cells,⁷ but the most efficacious dose among older nursing home residents has not been identified. The goal of this study was to identify the optimal dose of cranberry capsules that reduced the incidence of bacteriuria plus pyuria over one month.

METHODS

This study was a pilot double-blind, randomized, placebo-controlled trial of 3 cranberry capsules daily (108mg PAC), 2 cranberry plus one placebo capsule daily (72mg PAC), 1 cranberry plus two placebo capsules daily (36mg PAC), and 3 placebo capsules daily for 30 days. The primary outcome was episodes of bacteriuria plus pyuria at 7, 14, 21, and 28 days of cranberry capsule treatment. Participants were stratified by presence or absence of baseline bacteriuria with 20 participants randomized by strata to each arm of the study.

Cranberry and placebo capsules were donated by the manufacturer. Urine cultures and urinalyses were collected at baseline and then on a weekly basis for 4 weeks (total = 5 specimens). No additional follow-up was performed after the completion of cranberry capsule treatment. Bacteriuria was defined as >100,000 colony forming units per milliliter (cfu/ml) of any bacteria. Pyuria was defined as any white blood cells seen on microscopic urinalysis.⁵ Inclusion criteria were: 1)female; 2)history of UTI; 3)age≥65 years; 4)long term residence; 5)English speaking. Exclusion criteria included: 1)total incontinence; 2)warfarin therapy;⁸ 3)<4 weeks residence; 4)chronic indwelling catheter; 5)terminal prognosis; 6)antibiotic therapy; 7)kidney stones; 8)dialysis; 9)cranberry therapy; and 10)cranberry allergy. This study was approved by the Yale Human Investigation Committee. Eligible residents or surrogates provided written consent. SAS 9.22 (SAS Institute, Cary NC, 2009) statistical software was used to analyze the data.

RESULTS

In 11 homes, 1929 residents were screened; 1381 residents did not meet inclusion criteria (no history of UTI n=665[48.2%], male n=479[34.7%], short-term rehabilitation n=163[11.8%], non-English speaking n=38[2.8%], age < 65 years n=36[2.6%]). Of 548 remaining residents, 308 residents met exclusion criteria (totally incontinent n=104[33.8%], warfarin use n=72[23.4%], residence < 4 weeks n=31[10.1%], chronic indwelling catheter n=23[7.5%], discharged n=20[6.5%], terminal n=16[5.2%], antibiotic therapy n=14[4.5%], kidney stones n=9[2.9%], cranberry therapy n=7[2.3%], administrative decision n=6[1.9%], dialysis n=5[1.6%], cranberry allergy n=1[0.3%]) and 240 residents were eligible; 90 residents consented (37.5% consent rate), and 80 residents enrolled (10 participants met an exclusion criterion prior to enrollment). Demographics of the 80 participants included: 98% white(n=78), mean age 89.2 years(S.D., 7 years), and mean number of comorbidities 4.1(S.D., 1.7). Most participants were totally dependent in bathing(54%) and had some bowel(67%) and bladder(76%) incontinence. Of 80 baseline urine cultures, 1 had no growth, 8 had ≤100,000 cfu/ml, 41 had >100,000 cfu/ml, and 30 had mixed flora (3 or more organisms). The placebo, one, and three capsule groups each had 10 of 20 participants with >100,000 cfu/ml; the two capsule group had 11 of 20 participants with >100,000 cfu/ml on baseline urine culture. Of 80 baseline urinalyses requested, 73 were obtained: 11 had no pyuria and 62 had pyuria. Of 320 urine specimens that should have been collected, 302(94%) urine cultures and 294(92%) urinalyses were obtained. Results of bacteriuria plus pyuria by cranberry capsule group are provided in Table 1. These results represent 4 weekly follow-up urines obtained per participant while consuming cranberry capsules.

Table 1.

Cross-classification of cranberry capsule dose and presence of bacteriuria plus pyuria

Treatment Group	E.coli bacteriuria plus pyuria	Other bacteriuria plus pyuriaa	Not growthb	Total
No cranberry capsules	33 (43.4%)	5 (6.6%)	38 (50.0%)	76
One cranberry capsule	29 (40.3%)	4 (5.6%)	39 (54.2%)	72
Two cranberry capsules	23 (29.9%)	10 (13.0%)	44 (57.1%)	77
Three cranberry capsules	25 (34.3%)	12 (16.4%)	36 (49.3%)	73
Total	110	31	157	298c

^aOther bacteriuria plus any WBCs is >100,000 cfu/ml of a pathogen other than E.coli plus any WBCs. For no cranberry capsules, 3 were Proteus and 2 were Klebsiella species. For one cranberry capsule, 1 was Proteus and 3 were Klebsiella species. For two cranberry capsules, 2 were Proteus species, 1 Enterococcus, 4 beta-hemolytic Streptococci, 2 viridans Streptococci, and 1 Morganella morganii. For three cranberry capsules, 4 were Klebsiella, 3 Enterococcus, 4 Citrobacter freundii, and 1 coagulase negative Staphylococci.

^bNot growth includes no growth, growth < 100,000, growth >100,000 but no WBCs, and mixed flora. Of 157 not growth, 19 were no growth, 23 were <100,000, 4 were growth >100,000 but no WBCs, and 111 were mixed flora.

^cThere were 22 missing either urine culture or urinalysis of the 320 total expected samples.

Abbreviation: WBC (white blood cell)

DISCUSSION

This study showed a dose-dependent trend toward decrease in bacteriuria plus pyuria, particularly with E.coli, among female nursing home residents ingesting cranberry capsules over one month. Previous studies in older patients were conducted using cranberry juice,^{5, 9} and studies of cranberry capsules are lacking. In this study, E.coli bacteriuria was reduced and is consistent with the mechanism of PAC, ⁷ but bacteriuria with other pathogens did not show this same pattern of results. E.coli accounts for about 50% of uropathogens in nursing home residents,¹⁰ and reduction in bacteriuria may reduce treatment for UTI. Since the effect of two and three capsules was comparable and to reduce capsule burden, further investigation of two cranberry capsules daily in nursing home residents is warranted to determine if the reduction of E.coli bacteriuria is sustained over a longer period of time and whether it impacts clinical outcomes related to UTI (e.g., hospitalization, antibiotic therapy for UTI).