Research Article
Austin J Endocrinol Diabetes. 2016; 3(1): 1036.
Significance of the Dipstick in the Screening of Asymptomatic Bacteriuria in Diabetic Patients
Chemsi H*, Chadli M, Moutaouakkil Y, Sekhsokh Y and Elouennass M
¹Department of Bacteriology, Mohammed V Military Teaching Hospital Mohamed V-university, Morocco
²Research Laboratory and Biosafety, Mohammed V Military Teaching Hospital Mohamed V-university, Morocco
*Corresponding author: Hicham Chemsi, Department of Bacteriology, Mohammed V Military Teaching Hospital Mohamed V-university, Résidence Boustane ImmC3 Appt 12 Nahda. Rabat, Morocco
Received: November 26, 2015; Accepted: January 14, 2016; Published: January 18, 2016
Abstract
The urinary tract infections are very common and represent an important part of the workload in clinical microbiology laboratories. The aim of this study is assessing the interest of the Urine dipstick (BU) in excluding asymptomatic bacteriuria among diabetic patients, forming a significant part of consultants, and compare it with the reference examination that constitutes the cytobacteriological urine examination (urinalysis) for good decision making about the maintenance or removal of systematic ECBU. This is a prospective study in the laboratory of bacteriology at Mohamed V Military Hospital Instruction in Rabat (HMIMV), spread over six months from May to October 2012. The study included patient’s external diabetics, consultants and cooperative, presenting at the bacteriology lab for a urine culture. Patients with urinary catheter, incontinent and without diabetes were excluded from the study. All history (especially diabetes) was noted. The urine sample was taken as sterile as possible on which are made both a urinalysis and a test for BU. Patients were included among 427 diabetic subjects, female gender represented 180 (42%) and male was 247 (58%). Outcomes reported 173 positive strips (40%) suggestive of bacteriuria with a predominance of positive results in 92 women (21%) than in men 81 (19%). The results of the urine cultures are reported 36 ECBU (soit9%) reported bacteriuria, 21 (58%) men and 15 (42%) in women. The incriminated germ in 72% of cases were Escherichia coli, 330 (77%) sterile urine culture and 61 (14%) contaminated urine. In case of diabetes, atypical or absence of clinical signs evocative of urinary infection, requires the use of a reliable screening technique. Dipstick, which has been proven the excellent performance, fulfills perfectly the criteria for a screening test for urinary infection.
Keywords: Dipstick; Asymptomatic bacteriuria; Diabetic patients
Introduction
The urinary tract infections are very common and represent an important part of the workload in clinical microbiology laboratories. Diabetic patient is at higher risk of infection. Those infections are predominant among women and most remains asymptomatic. The aim of this study is assessing the interest of the Urine dipstick (BU) in excluding asymptomatic bacteriuria among diabetic patients, forming a significant part of consultants, and compare it with the reference examination that constitutes the cytobacteriological urine examination (urinalysis) for good decision making about the maintenance or removal of systematic ECBU.
Patients and Methods
This is a prospective study in the laboratory of bacteriology at Mohamed V Military Hospital Instruction in Rabat (HMIMV), spread over six months from May to October 2012. The study included patient’s external diabetics, consultants and cooperative, presenting at the bacteriology lab for a urine culture. Patients with urinary catheter, incontinent and without diabetes were excluded from the study. All history (especially diabetes) was noted. The urine sample was taken as sterile as possible on which are made both a urinalysis and a test for BU.
Urine are analyzed visually, a microscopic examination and culture to uro-calibrated loop was made on in culture. A test for BU (type Insight Expert) was made for each patient presenting for a urine culture. Except for the presence of leukocyte esterase and nitrites, other markers of BU were not taken consideration. The reading is made visually in comparison with the color scale. Detecting nitrites or leukocyte even at trace levels judge the test strip as evocative of bacteriuria. A strip is considered negative if one detects neither leukocyte esterase or nitrites. The diagnostic characteristics (i.e., predictive value, sensitivity, specificity, likelihood ratios) were calculated.
The major decision rule was the negative predictive value with its 95% confidence interval. An estimation of the cost-effectiveness was based on the same group of patients and based on approximate cost data collected from the billing department of HMIMV. All data, including the profile of patients, the results of urine culture and BU were recorded and analyzed using Microsoft Excel 2007.
Results
Patients were included among 427 diabetic subjects, female gender represented 180 (42%) and male was 247 (58%). The average age was 59 years. The results of the test strips are shown in Table 1, the values for the semi-quantitative characteristics are described. Outcomes reported 173 positive strips (40%) suggestive of bacteriuria with a predominance of positive results in 92 women (21%) than in men 81 (19%). The results of the urine cultures are reported in Table 2 with 36 ECBU (soit9%) reported bacteriuria, 21 (58%) men and 15 (42%) in women. The incriminated germ in 72% of cases were Escherichia coli, 330 (77%) sterile urine culture and 61 (14%) contaminated urine.
DIPSTICK
LEUKOCYTE ESTERASE
negative
277
Traces
1
positive (+)
98
positive (++)
19
positive (+++)
32
Nitrites
negative ( absence of nitrites )
355
positive ( presence of nitrites )
72
Table 1: Semi-quantitative results of the test strip markers of urinary tract infection.
RESULTS OF THE STRIP
RESULTS ECBU
sterile
contaminated
bacteriuria
Total
positive
108
33
32
173
negative
222
28
4
254
Total
330
61
36
427
Table 2: Comparison of the results of test strips and urine culture.
The “leukocytes isolated” situation has not been encountered. Comparing the results of the dipstick to the reference examination that is the ECBU is figured on Table 2, briefed us on the diagnostic performance of the dipstick and the discordance between the two tests was measured four cases of false negative described in Table 3. This comparison is used to calculate the performance of the dipstick in Table 4. The effectiveness of each strategy is defined by the sensitivity. The ECBU strategy being the reference strategy, it sensitivity is 1. It is 0.89 for BU strategy. The unnecessary use of ECBU expressed by the cost of induced positive urinalysis for patients detected (ie x% Positive strips cost ECBU).
SEX
AGES
LEUKOCYTURIA
(/ ML)
HEMATURIA
(/ ML)
CULTURE
bacteriuria
CFU / mL
germ
1
male
27
104
105
E.coli
2
male
76
4.106
2.104
106
E.coli
3
male
65
104
105
K.pneumoniae
4
female
53
104
105
E.coli
Table 3: Description of cytological and bacteriological characteristics of false negative results.
LEUKOCYTE ESTERASES OR
POSITIVE NITRITES
NPV (%) (Negative Predictive Value)
98,2
PPV (%) (Positive Predictive Value)
22,9
Sensitivity (%)
88,9
Specificity (%)
67,3
LR-(Likelihood ratio)
0,16
LR+(Likelihood ratio)
2,7
Undetected bacteriuria (%)
1
ECBU can be saved (%)
59
Table 4: Characteristics diagnostic test strip for the detection of bacteriuria in 427 diabetic leukocyte esterase or nitrite positive.
Discussion
Diabetic and above all in cases of chronic poor glycemic control patient has a higher risk of infection [1]. Diabetes is a complicating factor for urinary infections [2], independent risk factor for pyelonephritis [3,4]. The urinary tract infections are up to four times most frequent in diabetic patients compared to non-diabetic patient with a prevalence of 30%, they are predominant in women with a prevalence of 64.6% [5,6].
This was to evaluate the benefit of BU in excluding the asymptomatic bacteriuria in diabetic patients by measuring the discrepancy between the BU and ECBU with less workload and a moderate cost. NPV (negative predictive value) is among others, the primary endpoint its interest lies both in the fact that it is inversely proportional to the number of false negatives and closely linked with the number of cases of the disease in population in question. The study sample has a prevalence of bacteriuria (9%) recovered quite close to the rate (13%) patients with diabetes in the bacteriology laboratory in 2009 [7]. When a negative result of urine dipstick (N LE and negative), the probability of predicting the absence of bacteriuria (NPV = 98%) is very efficient. As a result, the discrepancy between the BU and urine culture is negligible with a proportion of 1% undetected bacteriuria (i.e. 4 false negative results).
The testing reveals a high sensitivity (89%), the relative lack of specificity (71.4%) is due to the high frequency of false positives (FP = 108 or 29.5%). A weak positive predictive value (PPV = 23%) is related to both high number of FP to the relatively low prevalence of bacteriuria (9%) which makes the PPV and NPV drop increases. The values of the positive and negative likelihood ratio (LR + = 2.7 and LR = 0.16) express a strong diagnostic supply of dipstick [8]. The BU has identified almost all of bactériuriques patients, it reduces the daily workload (up to 59% of the urine may be excluded from the analysis at the laboratory of bacteriology).
The data in ICE ratio, considering the unit cost of both tests and unnecessary use ECBU show that BU is a cost-effective test. However, reducing the workload, the contribution of human resources and the number of false negatives with impacts on the management of undetected asymptomatic patients, wichare hard to quantify and were not considered in the assessment ICE ratio. The data of the literature report diabetic bacteriuria a variable frequency ranging from 5.7% to 32.5% [9]. BU has already been assessed previously in France, a first study reported a prevalence of bacteriuria 9%, carried out in a specialized department of Endocrinology on retrospective data, revealed a relatively high rate of false negatives (FN = 12) [10].
A further study interested in diabetic subjects (with bacteriuria rate of 19%) showed a lower VPN, these results are improved by combining an unusual marker: the test glucose [11-17]. Our study affirming the effectiveness of BU and prevent the systematic sending a sample to the bacteriological laboratory in the event of a negative test. Therefore, achieving the urinalysis in diabetic patients is recommended in the diagnosis of symptomatic UTI or asymptomatic bacteriuria confirmation (if positive urinary strip).
Conclusion
In case of diabetes, atypical or absence of clinical signs evocative of urinary infection, requires the use of a reliable screening technique. Dipstick, which has been proven the excellent performance, fulfills perfectly the criteria for a screening test for urinary infection.
References
- Moutschen M. [Alterations in natural immunity and risk of infection in patients with diabetes mellitus]. Rev Med Liege. 2005; 60: 541-544.
- AFSSAPS (Agence française de sécurité sanitaire des produits de santé). [AFSSAPS Practice recommendations for diagnosis and antibiotic therapy of adult community urinary tract infections]. Med Mal Infect. 2008; 38: 203-252.
- Scholes D, Hooton TM, Roberts PL, Gupta K, Stapleton AE, Stamm WE. Risk factors associated with acute pyelonephritis in healthy women. Ann Intern Med. 2005; 142: 20-27.
- Fünfstück R, Nicolle LE, Hanefeld M, Naber KG. Urinary tract infection in patients with diabetes mellitus. Clin Nephrol. 2012; 77: 40-48.
- Bertal Filali K, Fouad Z, Diouri A. Infection urinaire et diabète. Diabetes Metab. 2008; 34: 40-100.
- Renko M, Tapanainen P, Tossavainen P, Pokka T, Uhari M. Meta-analysis of the significance of asymptomatic bacteriuria in diabetes. Diabetes Care. 2011; 34: 230-235.
- Karunajeewa H, McGechie D, Stuccio G, Stingemore N, Davis WA, Davis TME. Asymptomatic bacteriuria as a predictor of subsequent hospitalization with urinary tract infection in diabetic adults: The Fremantle Diabetes Study. Diabetologia. 2005; 48: 1288-1291.
- Geerlings SE, Stolk RP, Camps MJL, Netten MP, Collet TJ, Hoepelman AIM. Risk Factors for Symptomatic Urinary Tract Infection in Women with Diabetes. Diabetes care. 2000; 23: 1737-1741.
- Wilson ML, Gaido L. Laboratory diagnosis of urinary tract infections in adult patients. Clin Infect Dis. 2004; 38: 1150-1158.
- Diyane K, El Mghari G, El Ansari N. Analyse de l’ECBU chez les diabétiques hospitalisés. Diabetes Metab. 2012; 38: 32-111.
- Devillé WL, Yzermans JC, van Duijn NP, Bezemer PD, van der Windt DA, Bouter LM. The urine dipstick test useful to rule out infections. A meta-analysis of the accuracy. BMC Urol. 2004; 4: 4.
- Girard R, Montclos M, Bournaud C, Orgiazzi J. Dépistage des bactériuries à l’admission chez les patients diabétiques: peut on abandoner les examens cytobacteriologiques urinaires systématiques? Médecine et maladies infectieuses. 2006; 36: 219-222.
- Garrabé E, Cavallo J-D. Outils du diagnostic biologique des infections urinaires nosocomiales (IUN): analyse critique. Médecine et Maladies Infectieuses. 2003; 33: 447-456.
- Maaroufi A. Infection urinaire chez le diabétique: épidémiologie et profil de sensibilité des bactéries isolées aux antibiotiques. thèse de doctorat en médecine, sous la direction de Yassine Sekhsokh. Faculté de Médecine et de Pharmacie de Rabat. 2009.
- Delacoura H, Francois N, Servonneta A, Gentile A, Roche B. Les rapports de vraisemblance: un outil de choix pour l’interprétation des tests biologiques. Immunoanal Biol Spéc. 2009; 24: 92-99.
- Bally F, Troillet N. [Urinary tract infection: a tailored diagnosis]. Rev Med Suisse. 2008; 4: 2145-2148.
- Gérome P, Foucher B, Prevesto J-M, Chevalier B, Cheminel V, Bigois L, et al. Evaluation d’une pratique professionnelle : la bandelette urinaire permet-elle d’exclure le diagnostic d’infection urinaire chez le patient diabétique adulte hospitalisé. Ann Biol Clin. 2009; 67: 219-223.
Download PDF