International Urology and Nephrology 24 (1), pp. 3 - - 1 0 (1992)
Nitrofurantoin versus Trimethoprim for Low-dose Long-term Prophylaxis in Patients with Recurrent Urinary Tract Infections A Prospective Randomized Study W. VAHLENSIECKJR., +'* M. WESTENFELDER+'** Departments of Urology, *University of M/Jnchen; **Maria Hilf Hospital Krefeld; +University of Freiburg, Germany (Accepted February 14, 1991) In a prospective randomized study 38 patients with recurrent urinary tract infections (rUTI) were included to take either 50 mg Nitrofurantoin (n -----19) or 50 mg Trimethoprim (n = 19) as low-dose long-term prophylaxis for half a year. Compliance was checked weekly by Bacillus subtilis spore test strips sent in by mail. The infection rate was reduced from more than three per patient year to 0.01. There were no significant differences between the two groups concerning the recurrence rate (Nitrofurantoin: one rUTI; Trimethoprim: three rUTI) or side effects. Under Nitrofurantoin treatment 3 symptomatic fungal infections occurred. Trimethoprim and Nitrofurantoin are equally suitable for low-dose long-term prophylaxis in rUTI. Surveillance of compliance gives important hints for failure of prophylaxis.
Introduction Since about 25 years low-dose long-term prophylaxis with antimicrobial agents (LP) is introduced for treatment of recurrent urinary tract infections (rUTI) [1, 10, 13, 14, 15, 16]. The long reservoir function of the bladder at night is protected by nightly ingestion of antibiotics or urinary tract disinfectants. O f the dosage u s e d i n symptomatic infections 1/4 to 1/8 is given orally after the last micturition before sleep [8, 16]. Thus, bacteria having invaded the urinary tract do not have a chance for reproduction during the long urine stay in the bladder [12]. Some antibiotics like Norttoxacin or Trimethoprim also reduce the facultative pathogenic flora of the vagina and colon [12, 13]. Several drugs (e.g. Trimethoprim (TRM)-sulfamethoxazole combined or as single drugs, Cotetroxazin, Nitrofurantoin (FUR), nalidixic acid, oxolinic acid, Norfloxacin, methenamine mandelate and -hippurate) have been used for this purpose successfully [1, 8, 12, 13, 14, t5, 16J. Quinolones and disinfectants freeing formol in the urinary bladder are only second choice because of side effects or resistance problems. Tetracyclines, penicillins and oral cephalosporins are not optimal for this purpose because of low urinary concentration, high primary resistance quota or resistance induction [2, 5]. Fifty mg T R M given orally results in vaginal fluid and urinary concentration values exceeding 4 mg/1 M I C for most relevant urinary pathogens [5, 12]. T R M 1"
VSP, Utrecht Akad~miai Kiad6, Budapest
4
Vahlensieck, Westenfelder: Nitrofurantoin vs. Trimethoprim
prevents vaginal colonization by diffusion in vaginal fluid at bactericidal concentrations [2, 10]. F U R does not produce resistance in the fecal flora [5, 10]. Efficacy of the antimicrobial agent, development of resistance, severity of the underlying disease and reliable controls can be judged at the outpatient clinic. Reliable compliance control is possible only by detection of antimicrobial activity in the urine by test strips used by the patients themselves [14, 16].
Material and methods
Efficacy and tolerability of LP with respect to patient compliance were investigated in a clinical study. From February 1987 to February 1988, 38 adult patients (36 women, 2 men) with more than three UTIs every year were included in a prospective randomized study. They were followed for six months or until LP was interrupted, e.g., by breakthrough infection, adverse reaction or termination of LP by lack of compliance. At the beginning of the study all patients were free of infection at least for three days. Nineteen patients (range 21-75 years, average 43 years, one man) ingested 50 mg Nitrofurantoin (FUR; Furadantin(R)RP, RShm Pharma, Germany) and 19 patients (range 20-55 years, average 35 years, 1 man) 50 mg Trimethoprim (TRM; Trimono(R)RP, R t h m Pharma, Germany) orally every evening after the last micturition before sleep. Known allergies to the test drugs, pregnancy, breast feeding, neuritis, liver insufficiency, renal insufficiency, porphyria, glucose-6-phosphate-dehydrogenase deficiency, StevensJohnson syndrome, Lyell syndrome, exfoliative dermatitis, thrombocytopaenia, leucopaenia, anaemia and no possibility for 6-month compliance were exclusion criteria. Before starting treatment curable morphological or functional alterations of the urinary tract were excluded (see Table 3) [8]. Beside the medication all patients were given standardized general instructions to prevent rUTI [8] : to void whenever there was a desire to do so, to void immediately after sexual intercourse, to void without straining, to change clothes after bathing and to wipe from the urethral opening towards the anus after bowel movements. Antimicrobial activity in the urine was evaluated weekly with a commercially available test strip (Micur-(R)BT, B6hringer, Mannheim, Germany) utilizing inhibition of the growth of Bacillus subtilis spores [6]. After urinating over it at home during the first micturition of the day the patients enclosed it in a plastic box and sent it to our Department by mail every week. The patients were unaware of the significance of this procedure. After 24-hour incubation at 37 ~ a red colour of the stix indicated bacterial growth, no antimicrobial activity in the urine and non-compliance, and a white colour no bacterial growth, antimicrobial activity and compliance with medication [6, 14]. Also before and monthly during LP clinical outcome and complete clean catch urine phase contrast microscopy and routine bacteriology were controlled at our outpatient clinic. When breakthrough infections or allergies occurred the treatment was discontinued. Statistical analyses were carried out by Z2 testing on the absolute figures [7]. International Urology and Nephrology 24, 1992
Vahlensieck, Westenfelder: Nitrofurantoin vs. Trimethoprim
5
Results
There were no significant differences between the two groups concerning age, previous bacterial species in the urine, anamnestic urological diseases and results of the diagnostic procedures before treatment (Tables 1, 2 and 3). In the F U R group 63 ~ of the patients and in the T R M group 58 ~ had had urological diseases other than rUTI in their history. E. coli had been the dominant organism during the previous UTI episodes before starting prophylaxis. One patient in each group had two different organisms as cause of rUTI (E. coli and Klebsiella pneumoniae in the F U R group and E. coli and Enterococcusfaecalis in the T R M group). In the F U R group seven patients (39 ~ ) already had had long-term prophylaxis (one FUR, one Cotrimoxazol, two Norfloxacin, three T R M with three rUTI) and in the T R M group five (26 70) (three Cotrimoxazol, two F U R with one allergy). Only five patients (26 ~ ) in the F U R and nine (47 70) in the T R M group completed the study (Table 4). The others dropped out because of bad compliance (48 ~/32 ~), breakthrough infection (5 70/16 ~), mycosis (16 70/0 70) or allergy (5 ~ / 5 70). One patient after two weeks with F U R (5 ~ ) had clinical urinary tract re-infection with mixed urinary culture and the prophylaxis was stopped. At that time the Micur(R)BT strip indicated positive compliance. In the T R M group three patients after two, one and four months developed urinary tract re-infection (two E. coli, one StaphyTable 1 Previous urologic diseases in patients with rUTI (n = 38) Nitrofurantoin group (n = 19)
Bartholinian abscess (urethrotomy) Urethral dilatation Urinary incontinence Bladder cancer Pelvic radiation therapy Neurogenic bladder Ureteroneocystostomy Double kidney Vesicoureteral reflux Anderson-Hynes operation Nephrectomy Urolithiasis Diseases Patients
Trimethoprim group (n = 19)
Complete g r o u p ( n ~ 38)
1
0
1
0 3 2
3 1 0
3 4 2
1 3 1 1 3
0 1 2 1 2
1 4 3 2 5
1 1 1
1 1 3
2 2 4
18 (1 x 4.3• 2) 12 (63 ~)
15 (4x 2) 11 (58 ~)
33 23 (61 ~)
International Urology and NephrolotTy 24, 1992
Vahlensieek, Westenfelder : Nitrofurantoin vs. Trimethoprim
Table 2 Previous bacterial species in patients with rUTI (n = 38) before low-dose long-term prophylaxis Nitrofurantoin Trimethoprim Complete group(n ~ 19) group(n = 19) group(n = 38) E. coli Staph. aureus Staph. epidermidis Klebsiella pneumoniae Streptococci Proteus mirabilis Enterocoecus faecalis Unknown
11 1 2 1 0 0 0 5
2species
(58%) (5%) (10%) ( 5 %)
(27%)
1 (5~)
10 0 1 1 1 1 1 5
(53%)
21 1 3 2 1 1 1 10
(5%) ( 5 %) (5~) ( 5 %) ( 5 %) (27~)
1 (5%)
(55%) (3%) (7%) ( 5 %) ( 3 ~) ( 3 %) ( 3 %) (26%)
2 (5%)
Table 3 Diagnostic results in patients with rUTI (n = 38) before low-dose long-term prophylaxis Nitrofurantoin Trimethoprim Complete group (n = 19) group(n = 19) group(n ~ 38) Kidney sonography normal pathologic Residual volume normal ( < 100 ml) Uroflowmetry normal pathologic Urethral calibration normal Voiding cystourethrogram normal pathologic Cystoscopy normal pathologic IVP normal pathologic
10/12 2/12
12/14 2/14
22/26 (85 %) 4/26 (15 %)
15/15
15/15
30/30
11/14 3/14
10/15 5/15
21/29 (72 %) 8/29 (28 %)
5/5
4/4
7/14 7/14
9/14 5/14
16/28 (57 %) 12/28 (43 %)
14/16 2/16
10/10 0/10
24/26 (92 %) 2/26 ( 8 %)
11/16 5/16
13/18 5/18
24/34 (71%) 10/34 (29 %)
International Uroloyy and Nephroloyy 24, 1992
9/9
Vahlensieek, Westenfelder : Nitrofurantoin vs. Trirnethoprim
Table 4 Follow-up of low-dose long-term prophylaxis in rUTI (n = 38) Nitrofurantoin group (n = 19)
Trimethoprim group (n = 19)
Complete
5 (26 ~ )
9 (47 ~ )
Therapy ended before time due to: Bad compliance Infection Fungal infection Side effects
9 I 3 I
6 3 0 1
(48 ~ ) (5~) (16~) (5%7
(32 ~ ) (16~) (0K) (5%)
l o c o c c u s e p i d e r m i d i s ) . A l l three b a c t e r i a l strains were resistant to T r i m e t h o p r i m .
One M i c u r ( R ) B T strip at t h a t time i n d i c a t e d positive c o m p l i a n c e , the o t h e r t w o patients d i d n o t send strips at the time o f infection. T h r e e p a t i e n t s o f the F U R g r o u p after five m o n t h s (two) a n d after one week d e v e l o p e d s y m p t o m a t i c mycosis o f v a g i n a a n d bowel, vagina, a n d vagina, bowel a n d m o u t h . A l l three h a d t a k e n their p r o p h y l a x i s which was c o n f i r m e d b y a white M i c u r ( R ) B T strip at the time o f infection. I n the T R M g r o u p n o mycosis o c c u r r e d (p < 0.05). I n each g r o u p one Table 5 Bacillus subtilis spore test strips for compliance check
in patients with rUTI (n = 38) Nitrofurantoin group (n = 19) Projected 19 • 23 weeks (6 months) Dropout due to: Allergy Mycosis Recurrent UTI Bad compliance
437
437
23 29 18 159
Trimethoprim group (n = 19)
(5~) (7~) (4~) (36~)
19 ( 4 ~ ) 0 41 ( 9 ~ ) 81 (19 ~)
Weeks controlled
208/437 (48 ~)
296/437 (65 ~ )
No strips sent in Strips sent in
65/208 (31 ~) 143/208 (69 ~ )
105/296 (35~) 191/296 (65 ~ )
White strips Red strips
110/143 (77~) 33/143 (23~)
179/191 (94~) 12/191 ( 6 ~ )
International Urology and Nephrology 24, 1992
8
Vahlensieck, Westenfelder: Nitrofurantoin vs. Trimethoprim
case of side effects (5 %) occurred (TRM : rash, positive cutaneous reaction, after one month, white Micur(R)BT strip; FUR: GI-problems, just after starting therapy, no Micur(R)BT strip). In the FUR group 23 (5 %) of 437 projected strips dropped out because of allergy, 29 (7%) because of mycosis, 18 (4%) because of rUTI and 159 (36%) because of bad compliance, with the patients not showing up again (Table 5). These numbers were 19 (4 %) (allergy), 0 (0 %) (mycosis), 41 (9 %) (rUTI) and 81 (19 %) (bad compliance), respectively, in the TRM group. In the FUR group during the remaining actual surveillance time of the study 208 strips were projected. Actually 143 strips (69%) were sent in. The individual patient range here was 0-100% with a mean of 60%. Of the 143 strips 110 (77%) were white after incubation indicating positive compliance with medication. This number varied from patient to patient with a range of 0-100 % and a mean value of 66 %. These numbers were 296 (projected strips during actual surveillance time), 191 (65 %) (number of strips sent in), range 0-100 %/mean 66 % (individual patients sent in quota), 179 (94%) white (positive compliance in strips sent in), patient range 0-100%/mean 85 % (positive compliance in sent in strips for individual patients) for the TRM group. With one rUTI in 19 patients during 208 weeks of surveillance follow-up we got a recurrence rate of 0.01 UTI per patient year in the FUR group. In the TRM group three rUTI occurred in 19 patients during 296 weeks of follow-up, thus having a rate of 0.03 UTI per patient year. For all patients of both groups the risk for rUTI while patients were on prophylaxis was calculated as 0.01 per patient years. The differences between the two groups were not statistically significant.
Discussion
Before starting LP, especially with FUR in further studies, mycosis should be ruled out by vaginal smear culture. Because we did not check the patients for mycosis before LP we cannot tell if the three symptomatic Candida mycoses in the FUR group were exacerbations of preexistent colonizations or new infections under antibiotic therapy. More than 90 % of rUTI are re-infections rather than relapses [5, 10, 12]. So we did not insist on a positive susceptibility testing of the study drug for previous UTI organisms. The bad compliance was due to moving, family doctor's advise or holidays. The detection limits of Micor(R)BT in urine are 2 mg/1 for Trimethoprim and 8 mg/l for Nitrofurantoin [3]. After up to 6 hours' sleep the urinary concentrations following ingestion of 50 mg Trimethoprim or Nitrofurantoin before going to bed exceed these values (TRM: average about 30 mg/l, FUR: 15 mg/1) [4, 11 ]. So Micur(R)BT is apt for compliance control with this study protocol. With the help of the Micur(R)BT test strips it was demonstrated in two patients that they had taken their medication when breakthrough re-infection started. The other two did not send in strips at onset of re-infection. Side effects under TRM LP occurs in 7 %-11%, but only 4 %-7 % actually took their medicaInternational Urology and Nephrolofly 24, 1992
Vahlensieclr Westenfelder : Nitrofurantoin vs. Trimethoprim
9
tion when reporting allergy [12, 13, 14, 15, 16]. Side effects under F U R LP occur in 0 - 1 4 % (9 % GI-symptoms, 5 % cutaneous reactions) [1, 2, 10]. Irreversible lung fibrosis or other severe side effects concerning CNS or liver are very rare under low-dose long-term prophylaxis [1, 2, 5]. In our study one patient had an allergy to T R M proven by cutaneous test and white Micur(R)BT strip, the other side effect, G I symptoms under F U R , could not be confirmed by a positive strip test. About 2/3 of the patients under F U R and about 5/6 of the patients under T R M actually being surveyed took their medication. Breakthrough infections during LP are usually regarded as failure of LP. They will occur with resistant organism or if the medication is not ingested. Although there was no statistical difference, the slightly higher recurrence rate (3/19) in the T R M group again indicates the wellknown problem of a higher resistance induction rate of T R M or Cotrimoxazol than Nitrofurantoin or Norfloxacin [9, 12]. In a recent study we found 10/13 (77 %) resistant E. coli in breakthrough infections under Cotetroxazol LP [10, 14]. O f the patients, 42 % in the T R M and 26 % in the F U R group always t o o k their medication as compared to 5% and 21%, respectively, who never t o o k their medication indicating the importance of compliance checks, to prove if noncompliance might be the cause of recurrence. The breakthrough infection rate of 11% ( F U R group 5 %, T R M group 16 %) lies within the limits reported in the literature (13 %) [2, 12]. Again in this study LP reduces the rate of U T I about 10-fold [16]. T R M and F U R are both suitable for LP in r U T I . Regular compliance surveillance with commercially available bacterial spore test strips helps to differentiate the causes of LP failure in resistant organism and low compliance. Also side effects can be correlated with actual medication incorporation.
Acknowledgement We thank G. A. Rose M. D., London, for revising the manuscript.
References 1. Bailey, R. R., Roberts, A. P., Gower, P. E., DeWardener, H. E.: Prevention of urinary tract infection with low close Nitrofurantoin. Lancet, 2, 1112 (1971). 2. Breithaupt, H. : Grundlagen der Chemoprophylaxe unter besonderer Beriicksichtigung yon Trimethoprim und Nitrofurantoin aus klinisch-pharmakologischer Sicht. Aktuel. Urol., 18, Suppl. 1, 2 (1987). 3. Brillinger, G. U., Steinhausen, R. L., Supka, R., Wieczorek, L. R., Wielinger, H.: Personal communication. 4. Furadantin, Wissenschaftliche Informationsschrift. RShm Pharma, Darmstadt. 5. Huland, H. : Chemoprophylaxe bei Frauen mit rezidivierenden Harnwegsinfektionen. Aktuel. Urol., 18, Suppl. 1, 19 (1987). 6. Kaltwasser, F., Kaltwasser, G., Banauch, D., Wieczorek, L.: Nachweis antibakterieller Substanzen im Harn mit Teststreifen. Miineh. Med. Wochenschr., 125, 694 (1983). 7. Kollegium Biomathematik. Springer, Berlin-Heidelberg-New York 1977. International Urology and Nephrology 24, 1992
I0
Vahlensieck, Westenfelder : Nitrofurantoin vs. Trimethoprim
8. O'Grady, F., Kelsey, I., McSherry, A., Cattell, W. R. : Long-term treatment of persistent or recurrent urinary tract infection with Trimethoprim-Sulfamethoxazole. J. Infect. Dis., 128, Suppl. $652 (1973). 9. Sch/ifer, V., Weidner, W., Schiefer, H. G., Knothe, H. : Resistenzentwicklung unter Chemoprophylaxe mit Trimethoprim und Cotrimoxazol bei der rezidivierenden, unkomplizierten E. coli Harnwegsinfektion der Frau. Aktuel. UroL, 18, Suppl. 1, 23 (1987). 10. Stamey, T. A., Condy, M., Mihara, G. : Prophylactic efficacy of Nitrofurantoin macrocrystals and trimethoprim-sulfamethoxazole in urinary infections. N. Engl. J. Med., 296, 780 (1977). 11. Trimono, Wissenschaftliche Informationsschrift. R6hm Pharma, Darmstadt. 12. Weidner, W., Weil3bach, L., Eickenberg, H.-U., Carl, P., Meyer, W. : Trimethoprim und Cotrimoxazol zur Prophylaxe der rezidivierenden unkomplizierten Harnwegsinfektionen der Frau. Urologe [B], 25, 124 (1985). 13. Westenfelder, M., Pelz, K., Frankenschmidt, A., Vahlensieck, W. : Klinische Prtifung der Effektivit/it und Vertr/iglichkeit von Norfloxacin in der Therapie komplizierter Harnwegsinfektionen und in der Langzeitprophylaxe rezidivierender Harnwegsinfektionen. Infection, 15, 20 (1987). 14. Westenfelder, M., Vahlensieck, W., Reinarz, U. : Efficacy of and compliance with lowdose long-term prophylaxis in patients with recurrent urinary tract infections. Proc. IVth Intern. Pyelonephritis Symposium, G~Steborg, 1986. University Press, Chicago 1987, pp. 422-425. 15. Westenfelder, M., Vahlensieck, W., Reinarz, U. : Patient compliance and efficacy of lowdose, long-term prophylaxis in patients with recurrent urinary tract infections. Proc. Vth Mediterranean Congress of Chemotherapy, Cairo 1986. Chemiotherapia, 6, Suppl. 2, 530 (1987). 16. Westenfelder, M., Vahlensieck, W., Reinarz, U. : Patientencompliance und Effektivit/it der antimikrobiellen Langzeitprophylaxe mit Niedrigdosen bei Patienten mit rezidivierenden Harnwegsinfektionen (rHWI). AktueL UroL, 18, 6 (1987).
International Urology and Nephrolooy 24, 1992
Nitrofurantoin versus Trimethoprim for Low-dose Long-term Prophylaxis in Patients with Recurrent Urinary Tract Infections A Prospective Randomized Study W. VAHLENSIECKJR., +'* M. WESTENFELDER+'** Departments of Urology, *University of M/Jnchen; **Maria Hilf Hospital Krefeld; +University of Freiburg, Germany (Accepted February 14, 1991) In a prospective randomized study 38 patients with recurrent urinary tract infections (rUTI) were included to take either 50 mg Nitrofurantoin (n -----19) or 50 mg Trimethoprim (n = 19) as low-dose long-term prophylaxis for half a year. Compliance was checked weekly by Bacillus subtilis spore test strips sent in by mail. The infection rate was reduced from more than three per patient year to 0.01. There were no significant differences between the two groups concerning the recurrence rate (Nitrofurantoin: one rUTI; Trimethoprim: three rUTI) or side effects. Under Nitrofurantoin treatment 3 symptomatic fungal infections occurred. Trimethoprim and Nitrofurantoin are equally suitable for low-dose long-term prophylaxis in rUTI. Surveillance of compliance gives important hints for failure of prophylaxis.
Introduction Since about 25 years low-dose long-term prophylaxis with antimicrobial agents (LP) is introduced for treatment of recurrent urinary tract infections (rUTI) [1, 10, 13, 14, 15, 16]. The long reservoir function of the bladder at night is protected by nightly ingestion of antibiotics or urinary tract disinfectants. O f the dosage u s e d i n symptomatic infections 1/4 to 1/8 is given orally after the last micturition before sleep [8, 16]. Thus, bacteria having invaded the urinary tract do not have a chance for reproduction during the long urine stay in the bladder [12]. Some antibiotics like Norttoxacin or Trimethoprim also reduce the facultative pathogenic flora of the vagina and colon [12, 13]. Several drugs (e.g. Trimethoprim (TRM)-sulfamethoxazole combined or as single drugs, Cotetroxazin, Nitrofurantoin (FUR), nalidixic acid, oxolinic acid, Norfloxacin, methenamine mandelate and -hippurate) have been used for this purpose successfully [1, 8, 12, 13, 14, t5, 16J. Quinolones and disinfectants freeing formol in the urinary bladder are only second choice because of side effects or resistance problems. Tetracyclines, penicillins and oral cephalosporins are not optimal for this purpose because of low urinary concentration, high primary resistance quota or resistance induction [2, 5]. Fifty mg T R M given orally results in vaginal fluid and urinary concentration values exceeding 4 mg/1 M I C for most relevant urinary pathogens [5, 12]. T R M 1"
VSP, Utrecht Akad~miai Kiad6, Budapest
4
Vahlensieck, Westenfelder: Nitrofurantoin vs. Trimethoprim
prevents vaginal colonization by diffusion in vaginal fluid at bactericidal concentrations [2, 10]. F U R does not produce resistance in the fecal flora [5, 10]. Efficacy of the antimicrobial agent, development of resistance, severity of the underlying disease and reliable controls can be judged at the outpatient clinic. Reliable compliance control is possible only by detection of antimicrobial activity in the urine by test strips used by the patients themselves [14, 16].
Material and methods
Efficacy and tolerability of LP with respect to patient compliance were investigated in a clinical study. From February 1987 to February 1988, 38 adult patients (36 women, 2 men) with more than three UTIs every year were included in a prospective randomized study. They were followed for six months or until LP was interrupted, e.g., by breakthrough infection, adverse reaction or termination of LP by lack of compliance. At the beginning of the study all patients were free of infection at least for three days. Nineteen patients (range 21-75 years, average 43 years, one man) ingested 50 mg Nitrofurantoin (FUR; Furadantin(R)RP, RShm Pharma, Germany) and 19 patients (range 20-55 years, average 35 years, 1 man) 50 mg Trimethoprim (TRM; Trimono(R)RP, R t h m Pharma, Germany) orally every evening after the last micturition before sleep. Known allergies to the test drugs, pregnancy, breast feeding, neuritis, liver insufficiency, renal insufficiency, porphyria, glucose-6-phosphate-dehydrogenase deficiency, StevensJohnson syndrome, Lyell syndrome, exfoliative dermatitis, thrombocytopaenia, leucopaenia, anaemia and no possibility for 6-month compliance were exclusion criteria. Before starting treatment curable morphological or functional alterations of the urinary tract were excluded (see Table 3) [8]. Beside the medication all patients were given standardized general instructions to prevent rUTI [8] : to void whenever there was a desire to do so, to void immediately after sexual intercourse, to void without straining, to change clothes after bathing and to wipe from the urethral opening towards the anus after bowel movements. Antimicrobial activity in the urine was evaluated weekly with a commercially available test strip (Micur-(R)BT, B6hringer, Mannheim, Germany) utilizing inhibition of the growth of Bacillus subtilis spores [6]. After urinating over it at home during the first micturition of the day the patients enclosed it in a plastic box and sent it to our Department by mail every week. The patients were unaware of the significance of this procedure. After 24-hour incubation at 37 ~ a red colour of the stix indicated bacterial growth, no antimicrobial activity in the urine and non-compliance, and a white colour no bacterial growth, antimicrobial activity and compliance with medication [6, 14]. Also before and monthly during LP clinical outcome and complete clean catch urine phase contrast microscopy and routine bacteriology were controlled at our outpatient clinic. When breakthrough infections or allergies occurred the treatment was discontinued. Statistical analyses were carried out by Z2 testing on the absolute figures [7]. International Urology and Nephrology 24, 1992
Vahlensieck, Westenfelder: Nitrofurantoin vs. Trimethoprim
5
Results
There were no significant differences between the two groups concerning age, previous bacterial species in the urine, anamnestic urological diseases and results of the diagnostic procedures before treatment (Tables 1, 2 and 3). In the F U R group 63 ~ of the patients and in the T R M group 58 ~ had had urological diseases other than rUTI in their history. E. coli had been the dominant organism during the previous UTI episodes before starting prophylaxis. One patient in each group had two different organisms as cause of rUTI (E. coli and Klebsiella pneumoniae in the F U R group and E. coli and Enterococcusfaecalis in the T R M group). In the F U R group seven patients (39 ~ ) already had had long-term prophylaxis (one FUR, one Cotrimoxazol, two Norfloxacin, three T R M with three rUTI) and in the T R M group five (26 70) (three Cotrimoxazol, two F U R with one allergy). Only five patients (26 ~ ) in the F U R and nine (47 70) in the T R M group completed the study (Table 4). The others dropped out because of bad compliance (48 ~/32 ~), breakthrough infection (5 70/16 ~), mycosis (16 70/0 70) or allergy (5 ~ / 5 70). One patient after two weeks with F U R (5 ~ ) had clinical urinary tract re-infection with mixed urinary culture and the prophylaxis was stopped. At that time the Micur(R)BT strip indicated positive compliance. In the T R M group three patients after two, one and four months developed urinary tract re-infection (two E. coli, one StaphyTable 1 Previous urologic diseases in patients with rUTI (n = 38) Nitrofurantoin group (n = 19)
Bartholinian abscess (urethrotomy) Urethral dilatation Urinary incontinence Bladder cancer Pelvic radiation therapy Neurogenic bladder Ureteroneocystostomy Double kidney Vesicoureteral reflux Anderson-Hynes operation Nephrectomy Urolithiasis Diseases Patients
Trimethoprim group (n = 19)
Complete g r o u p ( n ~ 38)
1
0
1
0 3 2
3 1 0
3 4 2
1 3 1 1 3
0 1 2 1 2
1 4 3 2 5
1 1 1
1 1 3
2 2 4
18 (1 x 4.3• 2) 12 (63 ~)
15 (4x 2) 11 (58 ~)
33 23 (61 ~)
International Urology and NephrolotTy 24, 1992
Vahlensieek, Westenfelder : Nitrofurantoin vs. Trimethoprim
Table 2 Previous bacterial species in patients with rUTI (n = 38) before low-dose long-term prophylaxis Nitrofurantoin Trimethoprim Complete group(n ~ 19) group(n = 19) group(n = 38) E. coli Staph. aureus Staph. epidermidis Klebsiella pneumoniae Streptococci Proteus mirabilis Enterocoecus faecalis Unknown
11 1 2 1 0 0 0 5
2species
(58%) (5%) (10%) ( 5 %)
(27%)
1 (5~)
10 0 1 1 1 1 1 5
(53%)
21 1 3 2 1 1 1 10
(5%) ( 5 %) (5~) ( 5 %) ( 5 %) (27~)
1 (5%)
(55%) (3%) (7%) ( 5 %) ( 3 ~) ( 3 %) ( 3 %) (26%)
2 (5%)
Table 3 Diagnostic results in patients with rUTI (n = 38) before low-dose long-term prophylaxis Nitrofurantoin Trimethoprim Complete group (n = 19) group(n = 19) group(n ~ 38) Kidney sonography normal pathologic Residual volume normal ( < 100 ml) Uroflowmetry normal pathologic Urethral calibration normal Voiding cystourethrogram normal pathologic Cystoscopy normal pathologic IVP normal pathologic
10/12 2/12
12/14 2/14
22/26 (85 %) 4/26 (15 %)
15/15
15/15
30/30
11/14 3/14
10/15 5/15
21/29 (72 %) 8/29 (28 %)
5/5
4/4
7/14 7/14
9/14 5/14
16/28 (57 %) 12/28 (43 %)
14/16 2/16
10/10 0/10
24/26 (92 %) 2/26 ( 8 %)
11/16 5/16
13/18 5/18
24/34 (71%) 10/34 (29 %)
International Uroloyy and Nephroloyy 24, 1992
9/9
Vahlensieek, Westenfelder : Nitrofurantoin vs. Trirnethoprim
Table 4 Follow-up of low-dose long-term prophylaxis in rUTI (n = 38) Nitrofurantoin group (n = 19)
Trimethoprim group (n = 19)
Complete
5 (26 ~ )
9 (47 ~ )
Therapy ended before time due to: Bad compliance Infection Fungal infection Side effects
9 I 3 I
6 3 0 1
(48 ~ ) (5~) (16~) (5%7
(32 ~ ) (16~) (0K) (5%)
l o c o c c u s e p i d e r m i d i s ) . A l l three b a c t e r i a l strains were resistant to T r i m e t h o p r i m .
One M i c u r ( R ) B T strip at t h a t time i n d i c a t e d positive c o m p l i a n c e , the o t h e r t w o patients d i d n o t send strips at the time o f infection. T h r e e p a t i e n t s o f the F U R g r o u p after five m o n t h s (two) a n d after one week d e v e l o p e d s y m p t o m a t i c mycosis o f v a g i n a a n d bowel, vagina, a n d vagina, bowel a n d m o u t h . A l l three h a d t a k e n their p r o p h y l a x i s which was c o n f i r m e d b y a white M i c u r ( R ) B T strip at the time o f infection. I n the T R M g r o u p n o mycosis o c c u r r e d (p < 0.05). I n each g r o u p one Table 5 Bacillus subtilis spore test strips for compliance check
in patients with rUTI (n = 38) Nitrofurantoin group (n = 19) Projected 19 • 23 weeks (6 months) Dropout due to: Allergy Mycosis Recurrent UTI Bad compliance
437
437
23 29 18 159
Trimethoprim group (n = 19)
(5~) (7~) (4~) (36~)
19 ( 4 ~ ) 0 41 ( 9 ~ ) 81 (19 ~)
Weeks controlled
208/437 (48 ~)
296/437 (65 ~ )
No strips sent in Strips sent in
65/208 (31 ~) 143/208 (69 ~ )
105/296 (35~) 191/296 (65 ~ )
White strips Red strips
110/143 (77~) 33/143 (23~)
179/191 (94~) 12/191 ( 6 ~ )
International Urology and Nephrology 24, 1992
8
Vahlensieck, Westenfelder: Nitrofurantoin vs. Trimethoprim
case of side effects (5 %) occurred (TRM : rash, positive cutaneous reaction, after one month, white Micur(R)BT strip; FUR: GI-problems, just after starting therapy, no Micur(R)BT strip). In the FUR group 23 (5 %) of 437 projected strips dropped out because of allergy, 29 (7%) because of mycosis, 18 (4%) because of rUTI and 159 (36%) because of bad compliance, with the patients not showing up again (Table 5). These numbers were 19 (4 %) (allergy), 0 (0 %) (mycosis), 41 (9 %) (rUTI) and 81 (19 %) (bad compliance), respectively, in the TRM group. In the FUR group during the remaining actual surveillance time of the study 208 strips were projected. Actually 143 strips (69%) were sent in. The individual patient range here was 0-100% with a mean of 60%. Of the 143 strips 110 (77%) were white after incubation indicating positive compliance with medication. This number varied from patient to patient with a range of 0-100 % and a mean value of 66 %. These numbers were 296 (projected strips during actual surveillance time), 191 (65 %) (number of strips sent in), range 0-100 %/mean 66 % (individual patients sent in quota), 179 (94%) white (positive compliance in strips sent in), patient range 0-100%/mean 85 % (positive compliance in sent in strips for individual patients) for the TRM group. With one rUTI in 19 patients during 208 weeks of surveillance follow-up we got a recurrence rate of 0.01 UTI per patient year in the FUR group. In the TRM group three rUTI occurred in 19 patients during 296 weeks of follow-up, thus having a rate of 0.03 UTI per patient year. For all patients of both groups the risk for rUTI while patients were on prophylaxis was calculated as 0.01 per patient years. The differences between the two groups were not statistically significant.
Discussion
Before starting LP, especially with FUR in further studies, mycosis should be ruled out by vaginal smear culture. Because we did not check the patients for mycosis before LP we cannot tell if the three symptomatic Candida mycoses in the FUR group were exacerbations of preexistent colonizations or new infections under antibiotic therapy. More than 90 % of rUTI are re-infections rather than relapses [5, 10, 12]. So we did not insist on a positive susceptibility testing of the study drug for previous UTI organisms. The bad compliance was due to moving, family doctor's advise or holidays. The detection limits of Micor(R)BT in urine are 2 mg/1 for Trimethoprim and 8 mg/l for Nitrofurantoin [3]. After up to 6 hours' sleep the urinary concentrations following ingestion of 50 mg Trimethoprim or Nitrofurantoin before going to bed exceed these values (TRM: average about 30 mg/l, FUR: 15 mg/1) [4, 11 ]. So Micur(R)BT is apt for compliance control with this study protocol. With the help of the Micur(R)BT test strips it was demonstrated in two patients that they had taken their medication when breakthrough re-infection started. The other two did not send in strips at onset of re-infection. Side effects under TRM LP occurs in 7 %-11%, but only 4 %-7 % actually took their medicaInternational Urology and Nephrolofly 24, 1992
Vahlensieclr Westenfelder : Nitrofurantoin vs. Trimethoprim
9
tion when reporting allergy [12, 13, 14, 15, 16]. Side effects under F U R LP occur in 0 - 1 4 % (9 % GI-symptoms, 5 % cutaneous reactions) [1, 2, 10]. Irreversible lung fibrosis or other severe side effects concerning CNS or liver are very rare under low-dose long-term prophylaxis [1, 2, 5]. In our study one patient had an allergy to T R M proven by cutaneous test and white Micur(R)BT strip, the other side effect, G I symptoms under F U R , could not be confirmed by a positive strip test. About 2/3 of the patients under F U R and about 5/6 of the patients under T R M actually being surveyed took their medication. Breakthrough infections during LP are usually regarded as failure of LP. They will occur with resistant organism or if the medication is not ingested. Although there was no statistical difference, the slightly higher recurrence rate (3/19) in the T R M group again indicates the wellknown problem of a higher resistance induction rate of T R M or Cotrimoxazol than Nitrofurantoin or Norfloxacin [9, 12]. In a recent study we found 10/13 (77 %) resistant E. coli in breakthrough infections under Cotetroxazol LP [10, 14]. O f the patients, 42 % in the T R M and 26 % in the F U R group always t o o k their medication as compared to 5% and 21%, respectively, who never t o o k their medication indicating the importance of compliance checks, to prove if noncompliance might be the cause of recurrence. The breakthrough infection rate of 11% ( F U R group 5 %, T R M group 16 %) lies within the limits reported in the literature (13 %) [2, 12]. Again in this study LP reduces the rate of U T I about 10-fold [16]. T R M and F U R are both suitable for LP in r U T I . Regular compliance surveillance with commercially available bacterial spore test strips helps to differentiate the causes of LP failure in resistant organism and low compliance. Also side effects can be correlated with actual medication incorporation.
Acknowledgement We thank G. A. Rose M. D., London, for revising the manuscript.
References 1. Bailey, R. R., Roberts, A. P., Gower, P. E., DeWardener, H. E.: Prevention of urinary tract infection with low close Nitrofurantoin. Lancet, 2, 1112 (1971). 2. Breithaupt, H. : Grundlagen der Chemoprophylaxe unter besonderer Beriicksichtigung yon Trimethoprim und Nitrofurantoin aus klinisch-pharmakologischer Sicht. Aktuel. Urol., 18, Suppl. 1, 2 (1987). 3. Brillinger, G. U., Steinhausen, R. L., Supka, R., Wieczorek, L. R., Wielinger, H.: Personal communication. 4. Furadantin, Wissenschaftliche Informationsschrift. RShm Pharma, Darmstadt. 5. Huland, H. : Chemoprophylaxe bei Frauen mit rezidivierenden Harnwegsinfektionen. Aktuel. Urol., 18, Suppl. 1, 19 (1987). 6. Kaltwasser, F., Kaltwasser, G., Banauch, D., Wieczorek, L.: Nachweis antibakterieller Substanzen im Harn mit Teststreifen. Miineh. Med. Wochenschr., 125, 694 (1983). 7. Kollegium Biomathematik. Springer, Berlin-Heidelberg-New York 1977. International Urology and Nephrology 24, 1992
I0
Vahlensieck, Westenfelder : Nitrofurantoin vs. Trimethoprim
8. O'Grady, F., Kelsey, I., McSherry, A., Cattell, W. R. : Long-term treatment of persistent or recurrent urinary tract infection with Trimethoprim-Sulfamethoxazole. J. Infect. Dis., 128, Suppl. $652 (1973). 9. Sch/ifer, V., Weidner, W., Schiefer, H. G., Knothe, H. : Resistenzentwicklung unter Chemoprophylaxe mit Trimethoprim und Cotrimoxazol bei der rezidivierenden, unkomplizierten E. coli Harnwegsinfektion der Frau. Aktuel. UroL, 18, Suppl. 1, 23 (1987). 10. Stamey, T. A., Condy, M., Mihara, G. : Prophylactic efficacy of Nitrofurantoin macrocrystals and trimethoprim-sulfamethoxazole in urinary infections. N. Engl. J. Med., 296, 780 (1977). 11. Trimono, Wissenschaftliche Informationsschrift. R6hm Pharma, Darmstadt. 12. Weidner, W., Weil3bach, L., Eickenberg, H.-U., Carl, P., Meyer, W. : Trimethoprim und Cotrimoxazol zur Prophylaxe der rezidivierenden unkomplizierten Harnwegsinfektionen der Frau. Urologe [B], 25, 124 (1985). 13. Westenfelder, M., Pelz, K., Frankenschmidt, A., Vahlensieck, W. : Klinische Prtifung der Effektivit/it und Vertr/iglichkeit von Norfloxacin in der Therapie komplizierter Harnwegsinfektionen und in der Langzeitprophylaxe rezidivierender Harnwegsinfektionen. Infection, 15, 20 (1987). 14. Westenfelder, M., Vahlensieck, W., Reinarz, U. : Efficacy of and compliance with lowdose long-term prophylaxis in patients with recurrent urinary tract infections. Proc. IVth Intern. Pyelonephritis Symposium, G~Steborg, 1986. University Press, Chicago 1987, pp. 422-425. 15. Westenfelder, M., Vahlensieck, W., Reinarz, U. : Patient compliance and efficacy of lowdose, long-term prophylaxis in patients with recurrent urinary tract infections. Proc. Vth Mediterranean Congress of Chemotherapy, Cairo 1986. Chemiotherapia, 6, Suppl. 2, 530 (1987). 16. Westenfelder, M., Vahlensieck, W., Reinarz, U. : Patientencompliance und Effektivit/it der antimikrobiellen Langzeitprophylaxe mit Niedrigdosen bei Patienten mit rezidivierenden Harnwegsinfektionen (rHWI). AktueL UroL, 18, 6 (1987).
International Urology and Nephrolooy 24, 1992
