infection
Nieuwkoop, C. van
Citation
Nieuwkoop, C. van. (2011, February 17). Towards new strategies in complicated urinary tract infection. Retrieved from
https://hdl.handle.net/1887/16504
Version: Corrected Publisher’s Version
License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the
University of Leiden
Downloaded from: https://hdl.handle.net/1887/16504
Note: To cite this publication please use the final published version (if applicable).
137
Pelvic floor dysfunction is not related with incidence or outcome of urinary tract infection in adults
Cees van Nieuwkoop 1, Chantal H.M. Bergkamp 1, Henk W. Elzevier 2, Rob C.M. Pelger 2, Jan W. van’t Wout 1,3 and Jaap T. van Dissel 1
1 Dept. of Infectious Diseases, Leiden University Medical Center, Leiden, The Netherlands
2 Dept. of Urology, Leiden University Medical Center, Leiden, The Netherlands
3 Dept. of Internal Medicine, Bronovo Hospital, The Hague, The Netherlands
Submitted for publication.
Chapter 9.1.
To determine the relation between pelvic floor dysfunction (PFD) and urinary tract infection (UTI), a case-control study was performed. Consecutive adults with febrile UTI were included.
Those with PFD were cases and those without PFD were controls. Of 299 patients, 50 (26%) of 190 women and 24 (22%) of 109 men had PFD. Recurrent UTI was documented in 36% of female cases and controls; in men these percentages were 4% and 12%, respectively. The clinical and microbiological outcome was comparable between cases and controls. PFD is neither associated with incidence of UTI nor with outcome of febrile UTI in adults.
ABSTRACT
138
Introduction
Pelvic floor dysfunction (PFD) is a general term for functional problems affecting the urinary, rectal and/or sexual function, and it is considered to be a common problem in female populations 1, 2. Ac- cording to data from the American National Health and Nutrition Examination Survey approximate- ly, 24% of adult women have symptoms compatible with PFD 3. The incidence of PFD is associated with increasing age, obesity and childbirth and given the changing demographics, it is supposed to increase 3, 4. Moreover, PFD is considered to contribute to the development of urinary tract infections (UTI) in young women and children 5-7. However, there is an overall lack of data on the topic of UTI and PFD in elderly women and men. The incidence of UTIs in men increases with age, but a possible relation with PFD has, to our knowledge, never been studied and thus remains unclear.
After presentation with complicated or recurrent UTI, urological evaluation of the urinary tract may be considered. This may include evaluation of the pelvic floor which can be done by a pelvic floor questionnaire 3, 8. To select those patients at risk for PFD, risk factors for PFD should be iden- tified in men and women presenting with UTI. Furthermore, when PFD and UTI are etiologically related, it is not unlikely that PFD has an effect on the outcome of febrile UTI 9. However, in con- trast to previous studies, we recently have reported that PFD is not a risk factor for UTI in a general adult population 10. To gain more insight in the possible relation between PFD and UTI, the aim of this study was to determine whether PFD is associated with previous UTI or adverse outcome of febrile UTI, separately for men and women.
Patients and Methods
A nested case-control study was conducted in a multicenter cohort study. Consecutive adult pa- tients presenting with community-onset febrile UTI were enrolled at 8 emergency departments of regional hospitals and 35 neighboring primary healthcare centers between October 2006 and December 2009, as has been described previously 10, 11.
Enrolled patients were asked to complete a validated Pelvic Floor Questionnaire (Pelvic Floor In- ventories Leiden, PelFls) 28-32 days after presentation with febrile UTI, except those with one of the following exclusion criteria: presence of indwelling urinary catheter, neurologic impairment, bedridden status, uro- or colostomy, pregnancy, recent pelvic surgery or inability to fulfill a ques- tionnaire due to cognitive impairment or inability to read Dutch.
All questionnaires were evaluated independently by two urologists (HWE, RCMP) blinded to all patient data except sex. The results were dichotomized into presence or absence of PFD. In case of inter-observer disagreement, the final outcome was based on consensus after re-evaluation and discussion. Patients with presence of PFD were considered as cases and patients with absence of PFD were considered as controls.
Baseline demographics, clinical and microbiological data were collected by certified research pelvic floor dysfunction is not related with incidence or outcome of uti
139
nurses or the clinical investigators by personal patient interview and using data from the medical record. Urine and blood cultures were taken before starting antimicrobial therapy and were assessed using standard microbiological methods.
Co-morbidity was defined as the known presence of an urinary tract disorder, diabetes melli- tus, malignancy, chronic obstructive pulmonary disease (COPD), cerebrovascular disease or immunocompromised state, for which the patient is prescribed medication and/or consults a hospital-based medical specialist. Malignancy was defined as the diagnosis of a malignancy in the past 5 years, except basal cell carcinoma of the skin. Urinary tract disorder was defined as any self-reported history of an urological abnormality (benign prostate hyperplasia (BPH), ureteral re-implantation, vesicoureteral reflux, bladder carcinoma, urethral stricture). Recurrent UTI was defined as the occurrence of two or more self-documented UTIs in the previous 6 months or three or more in the previous 12 months.
Bacteremia at presentation, fever duration, bacteriuria at 30 days of follow-up and relapse or recur- rence of UTI during 90 days of follow-up (assessed by self reportage), were measured to investi- gate the outcome of febrile UTI. Fever duration was dichotomized at > 2 days as this is considered to be associated with a complicated course.
Bacteremia and significant bacteriuria were defined as described previously 11.
Statistical analysis was performed using SPSS software (SPSS Inc, Chicago, III, version 17.0). De- scriptive analysis included means or percentages with 95% confidence intervals (CIs) or medians and ranges, where appropriate. Inter-observer agreement for the dichotomized PelFls question- naire was assessed using the Cohen kappa-test, in which a k-value of 0.41-0.60 corresponds to fair agreement, 0.61-0.80 to good agreement, 0.81-0.92 t o very good agreement and 0.93-1.00 to excellent agreement 12. For comparison of cases and controls, univariate analysis was performed using the Student’s t test or Mann-Whitney U test for continuous variables and Chi-square tests for categorical variables. Measures for association were expressed as odds ratios (ORs) for disease with their 95% CIs for categorical variables. In addition, a multivariate logistic regression model was done to adjust for the following potential confounders that may be correlated with PFD: age, par- ity, diabetes mellitus, immunocompromised state, urinary tract disorder, body mass index (BMI), COPD and malignancy. This included the evaluation of potential interaction terms between these factors. To calculate ORs, age was dichotomized at > 50 years as surrogate for postmenopausal state and BMI at > 30 kg/m2 reflecting obesity. Data were analyzed for men and women separately.
A two-tailed P value of less than 0.05 was considered to indicate statistical significance.
Results
Between October 2006 and December 2009, 515 patients with febrile UTI were enrolled of which 110 patients had an exclusion criterion for PelFls assessment (Figure 1). Of the remaining 405 patients, 106 were not evaluable because they were lost to follow-up, died, refused or for other reasons (Figure 1). The overall response rate was 74%.
The inter-observer agreement between the two urologists for interpreting the PelFls question- naires was very good with a kappa of 0.824 (p < 0.001).
140
299 PelFls evaluation 109 Men | 190 Women
Women Of the 299 patients, 190 were women of which 50 (26%) had PFD and considered as fe- male cases (Table 1A); 140 were controls. Female cases were older (age of >50 years: OR 1.7; 95%
CI: 0.8-3.4), had higher BMI (BMI > 30 kg/m2: OR 2.1; 95% CI: 0.9-5.0) and more frequently had a history of pregnancy (parity ≥1: OR 2.0; 95% CI: 0.8-4.9). Malignancy was present in 7 (14%) of cases compared to 4 (3%) in controls (OR: 5.5; 95% CI: 1.5-19.8); 1 of 7 versus 1 of 4 malignancies were in the pelvic floor area for cases and controls, respectively. Female cases and controls did not differ with respect to smoking habits, diabetes mellitus, cerebrovascular disease, COPD, immu- nocompromised state, urinary tract disorder and co-morbidity in general. The multivariate ORs for malignancy was 3.8 (95% CI: 0.9-16.5), while the multivariate ORs for the other variables were similar to the univariate results.
pelvic floor dysfunction is not related with incidence or outcome of uti
figure 1
225 Controls without Pelvic Floor Dysfunction 85 Men | 140 Women 74 Cases with
Pelvic Floor Dysfunction 24 Men | 50 Women
405 Patients did meet inclusion criteria for
PelFls assessment 515 Patients with febrile UTI
208 Men | 307 Women
106 No PelFls evaluation 38 Lost to follow-up 18 Refused participation 13 Died within 30 days 34 No response
3 Incomplete questionnaire 110 Exclusion criterion 39 Urinary catheter 24 Cognitive impairment 19 Inability to read Dutch
9 Neurologic impairment 7 Bedridden status 7 Urostomy/Colostomy 4 Recent pelvic surgery 1 Pregnancy Enrolment of participants (UTI: urinary tract infection; PelFIs: Pelvic Floor Inventories Leiden)
141
Table 1a.
142
Baseline characteristics, clinical outcome and associations with Pelvic Floor Dysfunction of 190 women presenting with febrile UTI
Characteristic Pelvic Floor No Pelvic Floor Univariate OR P
Dysfunction Dysfunction [95%CI]
n = 50 (26) n = 140 (74)
Demographics / Risk factors
Age > 50 years 36 (72) 85 (61) 1.7 [0.8-3.4] 0.154
Body Mass Index > 30 kg/m² 11/49 (22) 16/134 (12) 2.1 [0.9-5.0] 0.076
Smoking 12 (24) 32 (23) 1.1 [0.5-2.3] 0.869
Co-morbidity
Any a 25 (50) 54 (39) 1.6 [0.8-3.1] 0.159
Diabetes mellitus 4 (8) 20 (14) 0.5 [0.2-1.6] 0.251
Malignancy 7 (14) 4 (3) 5.5 [1.5-19.8] 0.004
Cerebrovascular disease 6 (12) 9 (6) 2.0 [0.7-5.9] 0.210
COPD 6 (12) 10 (7) 1.8 [0.6-5.2] 0.288
Immunocompromised state 4 (8) 7 (5) 1.7 [0.5-5.9] 0.436
Urinary tract Disorder b 5 (10) 9 (6) 1.6 [0.5-5.1] 0.407
Obstetric History
Parity ≥ 1 37/44 (84) 95/131 (73) 2.0 [0.8-4.9] 0.123
History of UTI
Recurrent UTI c 18 (36) 51 (36) 1.0 [0.5-1.9] 0.957
Any previous UTI 35 (70) 100 (71) 0.9 [0.5-1.9] 0.848
Outcome of febrile UTI
Bacteremia 11/46 (24) 28/130 (22) 1.1 [0.5-2.5] 0.739
Fever duration >2 days 11/45 (25) 44/121 (36) 0.6 [0.3-1.3] 0.171
Bacteriuria d at 30 days follow-up 7/47 (15) 22/134 (16) 0.9 [0.4-2.2] 0.806 Relapse or recurrent UTI at 90 days follow-up 10/45 (22) 21/135 (16) 1.1 [0.4-3.5] 0.305
Data are presented as number (%). Abbreviations: UTI, urinary tract infection ; OR: odds ratio ; CI, confidence interval ; COPD, chronic obstructive lung disease. a Any co-morbidity: urinary tract disorder, diabetes mellitus, malignancy, COPD, cerebrovascu- lair disease or immunocompromised state; b Urinary tract disorder: any history of anatomical urological abnormality; c Recurrent UTI: at least two UTIs in the previous 6 months or three in the previous 12 months; d Bacteriuria is defined as bacterial growth over 103 CFU/ml urine or a bacterial monoculture over 102 CFU/ml urine in the presence of pyuria.
A prior history of recurrent UTIs were present in 36% of both female cases and controls, while 72%
of the cases had any UTI in history compared to 73% of the controls. There was no association between PFD and outcome of febrile UTI in women with respect to presence of bacteremia, dura- tion of fever, bacteriuria after 30 days follow-up and relapse or recurrence of UTI during 90 days follow-up (Table 1A). Multivariate analysis did not change the odds ratios with respect to history of UTI or outcome of febrile UTI.
Men Of 109 men, 24 (22%) had PFD (cases) and 85 had no PFD (controls) (Table 1B). Cases and controls did not differ with respect to age, BMI, co-morbidity, smoking habits, diabetes mellitus, cerebrovascular disease, COPD and immunocompromised state. A urinary tract disorder was present in 12 (50%) of cases compared to 28 (33%) of controls (OR 2.0; 95% CI:.0.8-5.1), with BPH accounting for 11 of 12 cases (46% of total) and for 23 of 28 controls (27% of total).
pelvic floor dysfunction is not related with incidence or outcome of uti
Table 1b. 143
Baseline characteristics, clinical outcome and associations with Pelvic Floor Dysfunction of 109 men presenting with febrile UTI.
Characteristic Pelvic Floor No Pelvic Floor Univariate OR P
Dysfunction Dysfunction [95%CI]
n = 24 (22) n = 85 (78)
Demographics / Risk factors
Age > 50 years 23 (96) 74 (87) 3.4 [0.4-27.9] 0.225
Body Mass Index > 30 kg/m² 4/23 (17) 11/81 (14) 1.3 [0.4-4.7] 0.646
Smoking 4 (17) 22 (26) 0.6 [0.2-1.9] 0.350
Co-morbidity
Any 17 (71) 55 (65) 1.3 [0.5-3.6] 0.576
Diabetes mellitus 5 (21) 14 (16) 1.3 [0.4-4.2] 0.619
Malignancy 1 (4) 14 (16) 0.2 [0.0-1.8] 0.122
Cerebrovascular disease 2 (8) 15 (18) 0.4 [0.1-2.0] 0.267
COPD 5 (21) 15 (18) 1.2 [0.4-3.8] 0.722
Immunocompromised state 1 (4) 5 (6) 0.7 [0.1-6.3] 0.745
Urinary tract Disorder 12 (50) 28 (33) 2.0 [0.8-5.1] 0.126
BPH 11 (46) 23 (27) 2.3 [0.9-5.8] 0.080
History of UTI
Recurrent UTI 1 (4) 10 (12) 0.3 [0.0-2.7] 0.275
Any previous UTI 7 (29) 29 (34) 0.8 [0.3-2.1] 0.649
Outcome of febrile UTI
Bacteremia 5/22 (23) 17/81 (21) 1.1 [0.4-3.4] 0.860
Fever duration > 2 days 6/20 (30) 22/67 (33) 0.9 [0.3-2.6] 0.812
Bacteriuria 30 days follow-up 7/23 (30) 10/83 (12) 3.2 [1.1-9.7] 0.033
Relapse or recurrent UTI at 90 days follow-up 5/21 (24) 18/83 (22) 1.6 [0.7-3.6] 0.834 Data are presented as number (%). BPH: benign prostate hyperplasia. For other abbreviations and definitions see Table 1a.
The presence of recurrent UTI was less frequent in men with PFD, 4% versus 12% controls (OR 0.3; 95% CI: 0.0-2.7), as was any UTI in history (OR 0.8; 95% CI: 0.3-2.1). Culture of urine samples obtained 30 days after initial presentation with febrile UTI, were available in 23 (96%) of the cases and 83 (98%) of the controls. The presence of significant bacteriuria was more frequent in cases;
7 (30%) (all asymptomatic except one) for cases compared to 10 (12%) in controls of which two were symptomatic (OR 3.1; 95% CI: 1.1-9.7, p = 0.033). The presence of bacteremia at presentation, fever duration and percentage of relapse or recurrence of UTI during 90 days follow-up were similar between cases and controls (Table 1B). Multivariate analysis did not alter the results for men.
Discussion
In this study we investigated the relation between PFD and UTI. We found that PFD is common and equally affects adult women and men who presented with febrile UTI; 26% and 22% respectively.
Moreover, in both women and men we found PFD neither to be related with incidence of previous episodes of UTI nor with the clinical course of febrile UTI. This indicates that PFD is not a risk factor for UTI. In women, PFD may be related with older age, obesitas, malignancy and previous pregnancy as has been documented in previous studies 3, 13. In men, PFD was also correlated with older age and in addition BPH seems to be a risk factor. Comparison with other studies in men is not feasible as, to our knowledge, this is the first study investigating risk factors for PFD in men without recent pelvic surgery.
Our study has several strengths. First of all it is a prospective case-control study including a large number of patients reflecting routine clinical practice. Secondly, the pelvic floor function was assessed by a validated pelvic floor questionnaire and evaluated by two independent blinded urolo- gists who had a very good inter-observer agreement.
There may however also be some limitations. Using strict criteria for pelvic floor assessment, 20%
of the original population was excluded which may limit the generalizibility of our findings. How- ever, in routine urologic practice similar criteria are being used to judge the pelvic floor function to be non evaluable (e.g. due to use of urinary catheter). In addition, some selection bias might have occurred because of the response rate of 74%. However, the similarity between baseline charac- teristics of the non evaluable subjects compared to the evaluated subjects, argues against an im- portant bias (data not shown). To interpret our results a few considerations have to be addressed.
First, malignancy was found to be a risk factor for PFD in women. This association has not previ- ously been described and it could not be explained by a location of the malignancy in the pelvic floor area. A possible explanation might be the use of opiates in such patients leading to constipa- tion or urinary retention, both symptoms of PFD.
Secondly, men with PFD had asymptomatic bacteriuria more frequently than those without PFD, while they did not report more UTIs during 90 days of follow-up. Though treatment of asymptom- atic bacteriuria in men is not to be recommended, indeed it can be a sign of underlying chronic bacterial prostatitis (CPB) which may not become clinical manifest within 90 days 14. Moreover, 144
there is an association between prostatitis and PFD 15. Taken together, the relation between PFD and asymptomatic bacteriuria in men as we report here, seem to be overlapping features of CPB.
This is of interest as men with asymptomatic bacteriuria who suffer PFD may thus differ from those without PFD and they may benefit from antimicriobial treatment to relieve PFD symptoms.
Clearly, further studies are needed to evaluate this hypothesis. Furthermore, BPH was also corre- lated with PFD. While BPH leads to urodynamic findings similar to those present in women with pelvic floor hypertonia, the majority of BPH patients in this study did not have PFD 2, 6. Likely, the lower urinary tract symptoms associated with BPH are similar to those listed in the questionnaire and the noted PFD is just a feature of more severe BPH symptoms.
Acknowledgements We acknowledge the patients for their participation, the primary care phy- sicians and their staff for enrolment of patients and the co-investigators of the following partici- pating hospitals: Groene Hart Hospital, Gouda: T. Koster; Rijnland Hospital, Leiderdorp: N.M.
Delfos; Diaconessenhuis Leiden, Leiden: H.C. Ablij; Medical Center Haaglanden, the Hague;
E.M.S. Leyten; Spaarne Hospital, Hoofddorp: G.H. Wattel-Louis.
References
1. Haylen BT, de Ridder D., Freeman RM et al. An International Urogynecological Association (IUGA)/
International Continence Society (ICS) joint report on the terminology for female pelvic floor dysfunction.
Int Urogynecol J Pelvic Floor Dysfunct 2010; 21(1):5-26.
2. Everaert K, van Laecke E, de Muynck M, Peeters H, Hoebeke P. Urodynamic assessment of voiding dysfunction and dysfunctional voiding in girls and women. Int Urogynecol J Pelvic Floor Dysfunct 2000;
11(4):254-264.
3. Nygaard I, Barber MD, Burgio KL et al. Prevalence of symptomatic pelvic floor disorders in US women.
JAMA 2008; 300(11):1311-1316.
4. Wu JM, Hundley AF, Fulton RG, Myers ER. Forecasting the prevalence of pelvic floor disorders in U.S.
Women: 2010 to 2050. Obstet Gynecol 2009; 114(6):1278-1283.
5. Finer G, Landau D. Pathogenesis of urinary tract infections with normal female anatomy. Lancet Infect Dis 2004; 4(10):631-635.
6. De Paepe H, Hoebeke P, Renson C et al. Pelvic-floor therapy in girls with recurrent urinary tract infections and dysfunctional voiding. Br J Urol 1998; 81 Suppl 3:109-13.:109-113.
7. Hooton TM. Recurrent urinary tract infection in women. Int J Antimicrob Agents 2001; 17(4):259-268.
8. Voorham-van der Zalm PJ, Stiggelbout AM, Aardoom I et al. Development and validation of the pelvic floor inventories Leiden (PelFIs). Neurourol Urodyn 2008; 27(4):301-305.
9. Minardi D, d’Anzeo G, Parri G et al. The role of uroflowmetry biofeedback and biofeedback training of the pelvic floor muscles in the treatment of recurrent urinary tract infections in women with dysfunctional voiding: a randomized controlled prospective study. Urology 2010;75(6):1299-304.
pelvic floor dysfunction is not related with incidence or outcome of uti
145
10. van Nieuwkoop C, Voorham-van der Zalm PJ, van Laar AM et al. Pelvic floor dysfunction is not a risk factor for febrile urinary tract infection in adults. BJU Int 2010; 105(12):1689-1695.
11. van Nieuwkoop C, Bonten TN, van ’t Wout JW et al. Risk factors for bacteremia with uropathogen not cultured from urine in adults with febrile urinary tract infection. Clin Infect Dis 2010; 50(11):e69-e72.
12. Byrt T. How good is that agreement? Epidemiology 1996; 7(5):561.
13. MacLennan AH, Taylor AW, Wilson DH, Wilson D. The prevalence of pelvic floor disorders and their relationship to gender, age, parity and mode of delivery. BJOG 2000; 107(12):1460-1470.
14. Nicolle LE, Bradley S, Colgan R, Rice JC, Schaeffer A, Hooton TM. Infectious Diseases Society of America guidelines for the diagnosis and treatment of asymptomatic bacteriuria in adults. Clin Infect Dis 2005;40(5):643-54.
15. He W, Chen M, Zu X, Li Y, Ning K, Qi L. Chronic prostatitis presenting with dysfunctional voiding and effects of pelvic floor biofeedback treatment. BJU Int 2010;105(7):975-7.
146
