Thesis
Reference
Les analyses urinaires systématiques avant une arthroplastie totale sont-elles indispensables?
BOUVET, Cindy
Abstract
La recherche et l'éradication d'une bactériurie asymptomatique avant la mise en place d'une prothèse de hanche ou de genou est controversée, et, pourtant essentielle pour le chirurgien.
L'influence pré-opératoire d'une dose d'antibiotique est inconnue. Notre étude prospective sur 510 arthroplasties a relevé les données d'examens urinaires avant et 3 jours après la chirurgie ainsi que de questionnaires à 3 mois de l'opération envoyé au patient et à son médecin traitant. Après analyse des données, notre conclusion est que la recherche systématique d'une bactériurie asymptomatique en pré-opératoire est coûteuse et ne permet pas de prédire le pathogène en cause. Si une infection symptomatique se produit, une antibiothérapie ciblée permet d'empêcher l'urosepsis et la diffusion hématogène vers le site de l'arthroplastie.
BOUVET, Cindy. Les analyses urinaires systématiques avant une arthroplastie totale sont-elles indispensables?. Thèse de doctorat : Univ. Genève, 2013, no. Méd. 10716
URN : urn:nbn:ch:unige-309661
DOI : 10.13097/archive-ouverte/unige:30966
Section de médecine clinique Département de chirurgie
Service de chirurgie orthopédique et traumatologie de l’appareil locomoteur
Thèse préparée sous la direction du Professeur Pierre HOFFMEYER et du Docteur Ilker UCKAY
" LES ANALYSES URINAIRES SYSTEMATIQUES AVANT UNE ARTHROPLASTIE TOTALE SONT-ELLES INDISPENSABLES ?"
Thèse
présentée à la Faculté de Médecine de l'Université de Genève
pour obtenir le grade de Docteur en médecine par
Cindy BOUVET
de
Lauterbrunnen (BE) Thèse n°10716
Genève 2013
RESUME
La recherche et l’éradication d’une bactériurie asymptomatique avant la mise en place d’une prothèse totale de hanche ou de genou est controversée, mais est d’importance capitale pour le chirurgien. L’influence de l’antibiothérapie administrée en per opératoire sur une potentielle bactériurie n’est pas connue. Nous avons donc réalisé une étude prospective avec observation d’une cohorte de patients. Les analyses urinaires étaient réalisées un jour avant la chirurgie et 3 jours post-opératoires. Les patients avec une infection urinaire symptomatique ou porteur de sonde à demeure étaient exclus de l’étude. Le suivi après l’opération incluait un questionnaire envoyé au patient ainsi qu’à son médecin traitant 3 mois après l’opération. Sur les 510 arthroplasties (290 hanche ; 309 femmes ; moyenne d’âge 69 ans), 182 (36%) avaient une bactériurie majoritairement due à E. Coli. Les sédiments urinaires pré opératoires ont révélé 181 épisodes de leucocyturie. La majorité des patients (95%) a reçu en per opératoire une dose unique de cefuroxime 1.5 g IV. La durée médiane des sondes urinaires en post-opératoire était 0 jours (entre 0 et 5). Au troisième jour post-opératoire les analyses étaient perturbées avec 99 épisodes de leucocyturie et 208 de bactériurie. La bactériurie au troisième jour était différente de celle pré- opératoire chez 51% des patients, la microbiologie révélait une proportion plus élevée d’enterocoque. Seulement 25 patients (5%) ont développé une infection urinaire symptomatique durant la période de suivi de 3 mois et un tiers des pathogènes étaient différents de ceux retrouvé durant l’hospitalisation. Toutes les infections symptomatiques ont été traité avec des antibiotiques oraux. Il n’y a eu aucun cas de dissémination vers les prothèses. Le coût estimé pour les analyses laboratoires pré-opératoire était de 23’120 € alors que l’absence d’utilisation d’antibiotique pour éradiquer des bactériuries asymptomatiques ont permis une économie d’au moins 2000€. Nous avons conclus que l’utilisation de routine des examens urinaires avant une arthroplastie totale était coûteux et ne permettait pas de prédire le germe en cause d’une potentielle infection
urinaire. Si une infection symptomatique se manifeste, une antibiothérapie ciblée permet d’éviter le sepsis urinaire et la dissémination hématogène vers le site de l’arthroplastie.
ABSTRACT
The search and eradication of asymptomatic bacteriuria before elective hip and knee arthroplasty is controversial, but reflects widespread practice. The influence of perioperative antibiotic prophylaxis on the dynamics of bacteriuria is unknown. We conducted a prospective observational cohort study with urine analyses before and 3 days after hip and knee arthroplasty. Patients with symptomatic urinary infections or long-term urinary catheter carriage were excluded. Post-discharge surveillance included questionnaires to patients and their general practitioners at 3 months. Among 510 arthroplasties (290 hip arthroplasties; 309 women; median age 69 yrs.), 182 (36%) had bacteriuria, mostly due to E. coli. Urine analysis also revealed 181 episodes of leukocyturia. Almost all patients (95%) had received a single-dose perioperative prophylaxis of cefuroxime 1.5 g IV. Median duration of postoperative urinary catheter carriage was 0 days (range, 0-5). On postoperative day 3 urinary analysis was abnormal with 99 episodes of leukocyturia and 208 episodes of bacteriuria, respectively.
Day 3-bacteriuria was different from preoperative sampling in 51% of patients, and microbiology revealed a higher proportion of enterococci. Only 25 patients (5%) developed a symptomatic urinary tract infection during a 3 months follow-up period, and one-third of pathogens were unrelated to those found during hospitalization. All symptomatic infections were treated with oral antibiotics.
There was no seeding of joint prostheses. Estimated minimal laboratory costs for preoperative urinary analyses were € 23,120, while the estimated savings by withholding antibiotics for bacteriuria eradication were at least € 2000. We conclude that pre- or postoperative routine urine evaluation of asymptomatic bacteriuria among patients undergoing hip and/or knee arthroplasty is costly and only moderately predicts the pathogen of a potential urinary tract infection. If a symptomatic infection occurs, targeted individualized antibiotic therapy prevents urosepsis and hematogenous spread to the arthroplasty site.
TABLE DES MATIERES
INTRODUCTION………Page 1
PATIENTS AND METHODS………..Page 5
RESULTS………...Page 8
DISCUSSION………...Page 12
BIBLIOGRAPHY……….Page 16
TABLE 1- Epidemiology of urine pathologies before and three days after arthroplasty....Page 19
TABLE 2 - Group comparison stratified according to symptomatic urine tract infection...Page 20
TABLE 3 - English-language literature review addressing the futility of preoperative urine analysis for the search of asymptomatic bacteriuria before arthroplasty...Page 21
INTRODUCTION
La recherche et l’éradication des bactériuries asymptomatiques avant la mise en place d’une prothèse totale de hanche ou de genou est controversée. Beaucoup de centres réalisent un stick urinaire1, un sédiment ou des cultures 2-4 avec un traitement antibiotique systémique durant 3 à 10 jours si la culture d’urine est positive 5-7. D’un autre côté, beaucoup de chirurgiens orthopédistes ne font pas de dépistages de routine (conviction personnelle). Alors que certaines sociétés nationales comme l’American Academy of Orthopaedic Surgeons recommandent des analyses urinaires seulement pour des patients avec une histoire récente ou fréquente d’infections urinaires8, d’autres9, ou des publications à opinion 10 évitent de donner des directives définitives.
D’après une étude par questionnaire réalisée au Royaume-Uni deux tiers des médecins interrogés traiteraient une bactériurie asymtomatique avant une arthroplastie de genou, alors que 70%
d’entre eux n’avaient aucune preuve concernant leur décision11. Notre expérience clinique suggère que les investigations urinaires en pre opératoire n’altèrent pas l’incidence post opératoires des infections d’arthroplasties. De plus, malgré le fait que dans la pratique cette question est récurrente, on retrouve peu de publications sur le sujet. Quelques unes recommandent un dépistage pré opératoire2-4, 10
alors que d’autres comptent sur le manque de fiabilité des marqueurs des perturbations urinaires. L’incidence décrite des infections urinaires après arthroplastie varie entre 0.7%12 et 2.4%13 à 15%14,surtout dans les cas de rétentions urinaires15. Selon certains auteurs, 80% des cultures urinaires effectuées avant la mise en place d’une prothèse reviennent négatives. Cependant, malgré ce résultat réassurant et l’absence totale de dissémination hématogène au site opératoire, les auteurs continuent à réaliser des analyses urinaires avant une arthroplastie élective3.
De routine, l’éradication des bactériuries potentielles est associée à des coûts de laboratoire et d’antibiothérapie élevés16 propre à chaque centre ce qui ne reflète en rien l’échelle mondiale. Il
est donc nécessaire d’obtenir plus de données. Cependant, la majorité des études sont rétrospectives3, 12, 17, 18
, ne font pas la différences entre une infection symptomatique et une colonisation asymptomatique17, 18, ont des données seulement pré opératoire7, 19, ne font pas la différence entre les arthroplasties mises dans un contexte traumatique ou dégénératif4, 7, n’ont pas de suivi après la sortie de l’hôpital2, 19, 20
, les coûts engendrés2, 12, 17, 20
et la prescription d’antibiotique inutiles4, 7, 12, 20
. La prédiction d’une infection symptomatique face à des analyses urinaires positives chez un patient porteur d’une arthroplastie est quasiment inconnue.
Notre hypothèse est que les analyses urinaires avant une arthroplastie en électif chez des patients asymptomatiques sont inutiles, coûteuses et donnent un travail sans intérêt aux infirmières. De plus un résultat positif pourrait mener le chirurgien orthopédiste à prescrire des antibiotiques sans indications. Nous avons donc réalisé une étude prospective avec une dose d’antibiotique prophylactique en per opératoire pour évaluer notre hypothèse. Il est démontré dans la littérature25 l’utilité d’une prophylaxie antibiotique lors d’une opération pour diminuer les risques infectieux de façon globale. Cependant l’effet de l’antibiothérapie per-opératoire sur une potentielle bactériurie n’est pas connu, c’est pourquoi notre étude se focalise sur les risques d’infection urinaire.
INTRODUCTION
The search and eradication of asymptomatic bacteriuria before elective total hip and knee arthroplasty is controversial, but reflects widespread practice with a large range of different approaches in resource-rich countries. Many centers perform dipsticks1, sediments, or cultures2-4, and treat for 3 to 10 days with systemic antibiotics if the urine analysis is positive5-7. On the other hand, many orthopaedic surgeons probably do not routinely investigate the urine (personal communications). While some national societies such as the American Academy of Orthopaedic Surgeons recommend urine analysis only for people with a history of recent or frequent urinary infections8, others9, or opinion papers10 avoid definitive statements. According to a postal survey in the United Kingdom two-thirds of physicians would treat asymptomatic bacteriuria prior to knee arthroplasty, yet 70% of them were unable to cite any evidence for their decision11. Our personal clinical experience suggests that regardless the degree of concern, preoperative urine assessment does not alter the incidence of arthroplasty infections.. Furthermore, despite its widespread practice little has been published on this issue. Some papers address preoperative urine screenings2-4, 10, while most others rely on surrogate markers of urine disturbances.
Reported incidences of urinary infections post-arthroplasty vary from 0.7%12 and 2.4%13 to 15%14, especially when there is postoperative urinary retention15. According to others, 80% of pre-arthroplasty urine cultures may be negative owever, despite this reassuring finding and total absence of hematogenous arthroplasty infection of urinary origin, the authors continue to perform urine analysis before elective arthroplasty3.
The routine, or even occasional assessment of bacteriuria with subsequent eradication is very likely associated with major laboratory and antibiotic costs16 within one specific institution, to say nothing about worldwide. It is clear that more data are needed. Unfortunately, most studies
are retrospective3, 12, 17, 18
, do not differentiate between symptomatic infection and asymptomatic colonization17, 18, rely only on preoperative assessments7, 19, fail to differentiate between arthroplasty for traumatic and degenerative conditions 4, 7, and do not take into account post- discharge surveillance2, 19, 20, associated costs2, 12, 17, 20
or unnecessary antibiotic prescription4, 7, 12,
20. Lastly, the value of positive urine analyses in predicting later symptomatic infection in the arthroplasty population is almost unknown.
We hypothesized that any urine analysis prior to elective arthroplasty in asymptomatic patients is useless, costly and an important factor in promoting unnecessary nursing work. Moreover, any positive results may push orthopaedic surgeons to prescribe needless antibiotic therapy. We performed a prospective study with single-shot antibiotic prophylaxis to evaluate this issue. We find in the litteratur the proof of the anbiotic prophylaxis before surgery25 for global infections, but we have no data specifically on the urinary infection. This is why our study is about the anbiotic prophylaxis during surgery on the post operative urinary infections.
PATIENTS AND METHODS
Setting before the study
The Orthopaedic Surgery Service at the Geneva University Hospitals has 132 acute care beds, a dedicated Infectious Diseases consultant21, and conducts a registry for total hip and knee joint prostheses22. Perioperative antibiotic prophylaxis consists of a single parental dose of cefuroxime 1.5 g. In cases of (anticipated) severe allergy to cephalosporins, or MRSA (methicillin-resistant Staphylococcus aureus) body carriage, 1 g of vancomycin is the alternative. Anesthesiologists and nurses participate to avoid perioperative urinary catheterization, or to keep it for as short a time as possible13, 23. Our catheters are not impregnated with silver or any other antiseptic substances.
Before the study urine sediment was routinely performed to screen for leukocyturia or nitrites in asymptomatic patients. Surgery was postponed in symptomatic patients. If the urine analysis was positive, a microbiological culture was added. Pathogens were usually eradicated with systemic antibiotics according to the microbiological profile or patients intolerances and co-morbidities.
Antibiotics were usually prescribed by surgeons and administered by nurses for an average of 5 days.
No control cultures were performed. Quantitative cultures and antibiotic susceptibility testing were performed according to the Clinical and Laboratory Standards Institute’s recommendations24. One urine sediment (microscopic exam) costs €12.5, and one urinary culture €92 if positive or €32 if negative. Antibiotic-related costs were obtained from the hospitals pharmacy21. Self-medication of antibiotics without medical prescription is almost impossible in Switzerland.
Prospective observational study
We evaluated the feasibility of a 1:1 randomized controlled trial (urinary tract evaluation and antibiotic treatment vs. no evaluation and no treatment of asymptomatic cases). We concluded that it
would be impossible due to sample size requirements. At maximum, only 1% of all arthroplasties get infected25, and only 25% of those infected are hematogenous in nature of which at most 40% would be of urinary origin. Moreover, we would have to exclude all patients without urine disturbances or significant bacteriuria4. Hence, we calculated a necessary sample size of 2 x 50,979 arthroplasties for a superiority trial (α 0.05, power 80%, difference in infection risk 0.10%-0.15%) and roughly 2 x 200,000 for a non-inferiority trial, impossible to achieve for the whole of Switzerland, let alone for a single center. Therefore, we opted for a prospective cohort study embedded in our arthroplasty registry.
Objectives, definitions and exclusion criteria
The primary objective was the epidemiological assessment of symptomatic urinary tract infections among elective arthroplasty patients without antibiotic prescription for asymptomatic urinary tract colonization. Secondary objectives were: a) The concordance of true symptomatic infection pathogens among those sampled before and after surgery (with perioperative antibiotic prophylaxis);
b) The epidemiology of pathological urinary sediments before and after surgery, with the corresponding concordance; c) Cost issues related to urinary analyses; and d) Clinical variables associated with symptomatic infection.
Leukocyturia was defined as more than 16 leucocytes in a microscopic field, and asymptomatic bacteriuria as more than 10³ colony forming units/ml of culture. This is the minimal threshold recommended in the literature and daily practice for the microbiological diagnosis of urine infection6.
chronic urinary tract diseases such as cancer, patients with chronic indwelling urinary catheter carriage26, antibiotic medication for any reason during the preceding two weeks, dementia, and an expected hospital stay shorter than 5 days (young patients without co-morbidities).
Practical aspects
Between November 2011 and September 2012 all patients undergoing elective hip or knee arthroplasty on our service were invited to participate on a voluntary basis. On admission, urine sediment exams and urine cultures were sampled and repeated at Day-3 postoperatively. We chose Day-3 to be sure that there were no remaining effects of cefuroxime, our perioperative prophylaxis (half-life-time 1-1.5 h). Additionally, Day-3 was early enough to sample urine before the scheduled discharge of most patients. Patients were followed through 30 April 2013; e.g. until six months after the inclusion of the last patient. Post-discharge surveillance included free-of-charge questionnaires sent to patients and their general practitioners at 3 months postoperative. These questionnaires were identical, and asked for urinary tract symptoms and related antibiotic use. We additionally requested information on pathogens, if present, as well as medical confirmation of infection from the general practitioner. The local Ethics Committee approved both the Arthroplasty Registry (no. 08-057) and our nested study (no. 10-148).
With respect to the statistical analysis we performed group comparisons using Pearson χ2-test. P values ≤ 0.05 (all two-tailed) were considered significant. STATA™ software (9.0; College Station, USA) was used.
RESULTS
Study population
During the study period, 615 arthroplasties (386 hip and 229 knee) were performed. Of these, 105 were excluded for various reasons, as follows: Non-elective surgery (n=4); Under antibiotic medication on admission (n=3); Refusal of study participation or forgetting to sign before admission (n=80); Postoperative antibiotic administration for suspicion of pneumonia (n=8); Urothelial cancer (n=1); Patient refusal of urinary catheter removal when it was possible (n=1); Chronic indwelling catheter (n=1); and Short hospital stay or short follow-up time (n=7).
For final analysis there were 510 interventions (290 total hip and 220 total knee arthroplasties) in 504 patients (309 women (61%); mean age 69.1 years; range 16-97 yrs.) who met the study criteria and were followed for at minimum of three months post-discharge. Sixty-eight patients (13%) were immune-compromised for the following reasons: diabetes mellitus (n=45), severe renal insufficiency and dialysis (n=9), autoimmune diseases requiring immunosuppressive medication (n=5), advanced cirrhosis (n=2), and non-urological cancers (n=7). Perioperative urinary catheterization lasted for a median of 0 days (range, 0-5 days). Indeed, only after 14 interventions (3%) was urinary catheterization necessary for longer than 12 hours postoperatively .In 497 interventions a single-dose of cefuroxime was given as peri-operative prophylaxis, and in 11 interventions three doses.
Urine analyzes before and after surgery
nitrites (Table 1). Only 315 episodes (62%) yielded the same leukocyturia assessment (present/absent) before and after surgery, compared to 468 nitrite episodes (92%). Variables significantly associated with pathological sediment before surgery were female gender and age over 75 years (χ2-tests; both p=0.001), while the presence of immune suppression was not.
Regarding bacterial colonization, 182 (36%) positive preoperative and 208 (41%) positive postoperative cultures yielded 17 different organisms, of which the five most frequent were Escherichia coli (n=109), Enterococcus faecalis (n=66), Streptococcus agalactiae (n=13), Klebsiella pneumoniae (n=8), and Enterobacter spp (n=8). Many colonizations were polymicrobial, especially in the postoperative sampling (n=53) as compared to the preoperative sampling (n=24). The spectrum of cefuroxime prophylaxis covered 119 pre-operative samplings (119/182; 65%) but only 22 (22/208;
11%) of the postsurgical samplings, for which the difference was significant (χ2-test; p<0.0.001).
Hence, cefuroxime did not reduce quantitative urine colonization, but shifted the microbiology more towards the enterococci (χ2-test; p<0.0.001) and decreased leukocyturia and the urine nitrite counts (Table 1). Qualitatively, half of the urine cultures (191/sum of 182 + 208; 51%) yielded another pathogen within the four-day interval between the urine samples.
Costs
For the traditional urine assessment before the start of this study, the costs related to the urinary analysis (and cultures if sediment was pathologic) equaled €23,120. This could have been avoided by abandoning any urine analysis under all study conditions. We also added the cost- sparing effect of antibiotic prescription when we would not treat colonization. Based on internal hospital prices and the usual use of oral co-trimoxazole, amoxicillin or quinolones, we computed a savings sum of €2000. With a conservative calculation, abandoning preoccupation with preoperative urine analysis would probably save €25,000 a year (€50 per patient per 500 patients), workload not included.
Questionnaires
For 446 interventions (87%) at least one questionnaire was returned from either the patient or the general practitioner. While 28 patients (5%) reported some minor dysuria (based on descrption of symptoms), their physicians only confirmed 13 episodes of infection (2.5%). Thirty-two patients (6%) received antibiotic agents for an origin other than the urine during the post-discharge surveillance period.
Symptomatic infections
Twenty-five patients (5%) developed a symptomatic cystitis during the 3 month follow-up period, despite withholding all antibiotic treatment for pathological asymptomatic bacteriuria. There were no cases of pyelonephritis, urinary tract abscess or urosepsis. Causative pathogens were mostly E. coli
or seeding to the joint prosthesis. Table 2 shows the study population stratified according to presence or absence of symptomatic infection. The subgroups were equally divided, except for preoperative urine disturbances that were more often associated with later cystitis.
We were also interested how preoperative urine disturbances could predict later urinary infection.
The sensitivity, specificity, positive and negative predictive values of preoperative leukocyturia were 0.56, 0.66, 0.08, and 0.97. The corresponding results for preoperative nitrites were 0.20, 0.95, 0.18, and 0.96, whereas for a positive urine culture on admission they were less at 0.56, 0.57, 0.08, and 0.76, respectively (Table 2).
DISCUSSION
In our prospective cohort study of 510 elective arthroplasties with a three-month post-discharge surveillance, only 25 patients (5%) developed a symptomatic cystitis, while withholding all antibiotic treatment even for positive bacteriuria. Two-thirds of these symptomatic infections occurred post- discharge. Among these, two-thirds of the post-discharge pathogens were unrelated to those found during hospitalization. All symptomatic infections were successfully treated with oral antibiotics without recurrence or seeding to the prosthesis.
The low proportion of cystitis is in contrast to the high proportion of pathological sediments and positive urine cultures before and after surgery. Single dose perioperative antibiotic prophylaxis failed to reduce the number of postsurgical colonizations, but shifted the germ spectrum towards more enterococci. This is not surprising since enterococci are in vivo resistant to all cephalosporins, including cefuroxime. In other words, while many surgeons worry about the presence of potential Gram-negative bacteria in the urine before surgery, they ignore that most of them still hide there afterwards. Theoretically, the situation has become worse since we are now facing Gram-positive pathogens with a freshly implanted total joint prosthesis. Gram-positive bacteria such as staphylococci, streptococci and enterococci are the hallmark of orthopaedic infections, and not Gram-negatives25. In the literature, preoperative27 and postoperative28-30 bacteriuria may both occur with a frequency of 12%-18% depending on the duration of urine catheter carriage26. Of note, pathological sediment was associated with both female gender and elderly patients, populations already known to be at risk for cystitis9.
antibiotic eradication of asymptomatic urine colonization9. To infect a prosthetic joint the urine infection must first become bacteremic. This is a relatively rare event and concerns primarily lower urinary tract infections that have been neglected. While the epidemiological risk for ascension to the kidneys is 5%31, only one cystitis episode among 235 urinary catheter carriers was bacteremic in a prospective study32. Moreover, placebo-controlled trials regularly show that women are able to cure symptomatic cystitis to at least 28% to 58%without antibiotics within a few days33, 34, suggesting that there is normally sufficient time to prescribe antibiotics before a symptomatic cystitis worsens.
Among the orthopaedic population the majority of publications in our literature review (Table 3) agrees with our findings. Koulouvaris et al. retrospectively investigated the relationship between a perioperative urinary tract infection and prosthesis infection among 20,000 arthroplasty patients,and found no association7. Other groups as well failed to establish a relationship between asymptomatic bacteriuria and arthroplasty infections2, 35. A retrospective analysis of our own arthroplasty registry from 1996-2008 revealed 73 arthroplasty infections, none of which was attributed to a urinary infection in the postoperative period. At the same time these patients were exposed to urinary colonization for a median duration of 295 days, corresponding to a cumulative burden of 120,000 patient-days36. Recently, in 471 patients researchers investigated if antibiotic therapy for asymptomatic bacteriuria would prevent hematogenous seeding to freshly implanted hip prostheses4. Above a cut-off of 105 colonies/ml in preoperative urine cultures they randomized patients to receive or not to receive systemic antibiotics for 7 days, and found no hip joint infections of urinary origin.
They stated that one would need to treat 25,000 asymptomatic bacteriurias in order to prevent one arthroplasty infection. However, as noted above their study does not by far reach the required sample size to adequately evaluate this question with a randomized controlled trial. The same group published a similar study three years previously with the same conclusion, that an antibiotic therapy
for 7 days for asymptomatic bacteriuria is not associated with hematogenous seeding to the prostheses5.
Our study has some limitations, as follows: (i) It originates from a single urban institution in a high- income country, aspects that might limit extrapolation of its findings. For example, our costs calculations are based on our clinic with its own costs and screening politics (urine sediment examination). Outside Geneva, where urine analyses would rely only on dipsticks3, 37 or are performed occasionally, laboratory costs could be significantly reduced; (ii) We excluded transplant patients, patients with urothelial or prostatic cancers, arthroplasties for traumatic disorders, and patients with chronic urinary catheter carriage, and thus the results do not apply to these patient groups; (iii) Post-discharge surveillance was based on questionnaires sent to patients and their general practitioners. Patient response relied on subjective interpretation, whereas the medical diagnosis of (past) symptoms was supported by laboratory results in the majority of cases. However, both could not be verified by the study team. To cite one extreme, according to the geriatric literature only 19% of the institutionalized elderly truly have a urinary tract infection despite the presence of fever, urogenital symptoms and significant bacteriuria38; (iv) In thirteen percent of cases the questionnaire was missing, but this is considered a more than acceptable return in scientific surveys;
(v) We excluded patients under antibiotic therapy and those reporting treatment for a urinary tract infection within two weeks before admission. However, the time delay beyond which there is no influence of antibiotic consumption on urinary tract colonization is unknown. For example, even short administrations of systemic antimicrobial agents for respiratory tract infections two months
In conclusion, if urinary catheters are removed as soon as possible preoperative or postoperative routine urine evaluation of asymptomatic arthroplasty patients is costly and potentially antibiotic- promoting. Preoperative urine analyzes only moderately predict the pathogen of a later infection.
Pathological sediment only confirms an already known increased risk for infection in women and patients over 75 years of age. Moreover, perioperative antibiotic prophylaxis and surgery alter the microbiology of half the urine colonizations, but do not eliminate them. If symptomatic infection occurs, there is sufficient time for a targeted individualized antibiotic therapy preventing urosepsis and hematogenous spread to joint prostheses18. Consequently, we have abandoned urine sampling before arthroplasty with the exception of transplant patients, patients with urinary tract pathologies and those with chronic indwelling catheter carriage. Based on conservative estimates we would spare
€25,000 per year per 500 arthroplasty patients
BIBLIOGRAPHY
1. Singh D, Roberts C, Bentley G. Urinalysis before joint arthroplasty. To dipstick or not?
That is the question. Annals of the Royal College of Surgeons of England 1998;80(4):300.
2. Glynn MK, Sheehan JM. The significance of asymptomatic bacteriuria in patients undergoing hip/knee arthroplasty. Clinical orthopaedics and related research 1984(185):151-4.
3. Clement S, Young J, Munday E. Comparison of a urine chemistry analyser and microscopy, culture and sensitivity results to detect the presence of urinary tract infections in an elective orthopaedic population. Contemporary nurse 2004;17(1-2):89-94.
4. Cordero-Ampuero J, Gonzalez-Fernandez E, Martinez-Velez D, Esteban J. Are Antibiotics Necessary in Hip Arthroplasty With Asymptomatic Bacteriuria? Seeding Risk With/Without Treatment. Clinical orthopaedics and related research. Epub ahead of print 2013 Feb 21.
5. Martinez-Vélez D G-FE, Cordero-Ampuero J, Casa de Pantoja V. Asymptomatic urinary tract infection in patients undergoing elective hip and knee arthroplasty. J Bone Joint Surg Br 2010;92(85):12-23.
6. David TS, Vrahas MS. Perioperative lower urinary tract infections and deep sepsis in patients undergoing total joint arthroplasty. The Journal of the American Academy of Orthopaedic Surgeons 2000;8(1):66-74.
7. Koulouvaris P, Sculco P, Finerty E, Sculco T, Sharrock NE. Relationship between perioperative urinary tract infection and deep infection after joint arthroplasty. Clinical orthopaedics and related research 2009;467(7):1859-67.
8. American Academy of orthopaedic Surgeons. OrthoInfo.
http://orthoinfoaaosorg/topiccfm?topic=a00389 2011.
9. 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.
10. Rajamanickam A, Noor S, Usmani A. Should an asymptomatic patient with an abnormal urinalysis (bacteriuria or pyuria) be treated with antibiotics prior to major joint replacement surgery? Cleveland Clinic journal of medicine 2007;74 Suppl 1:S17-8.
11. Finnigan TKP BM, Shepard GJ. Asymptomatic bacteriuria prior to arthroplasty: how do
13. Iorio R, Whang W, Healy WL, Patch DA, Najibi S, Appleby D. The utility of bladder catheterization in total hip arthroplasty. Clinical orthopaedics and related research 2005(432):148-52.
14. Yang K, Yeo SJ, Lee BP, Lo NN. Total knee arthroplasty in diabetic patients: a study of 109 consecutive cases. The Journal of arthroplasty 2001;16(1):102-6.
15. Petersen MS, Collins DN, Selakovich WG, Finkbeiner AE. Postoperative urinary retention associated with total hip and total knee arthroplasties. Clinical orthopaedics and related research 1991(269):102-8.
16. Lawrence VA, Gafni A, Gross M. The unproven utility of the preoperative urinalysis:
economic evaluation. Journal of clinical epidemiology 1989;42(12):1185-92.
17. Kumar P, Mannan K, Chowdhury AM, Kong KC, Pati J. Urinary retention and the role of indwelling catheterization following total knee arthroplasty. Int Braz J Urol 2006;32(1):31-4.
18. Pulido L, Ghanem E, Joshi A, Purtill JJ, Parvizi J. Periprosthetic joint infection: the incidence, timing, and predisposing factors. Clinical orthopaedics and related research 2008;466(7):1710-5.
19. Ollivere BJ, Ellahee N, Logan K, Miller-Jones JC, Allen PW. Asymptomatic urinary tract colonisation predisposes to superficial wound infection in elective orthopaedic surgery.
International orthopaedics 2009;33(3):847-50.
20. Donovan TL, Gordon RO, Nagel DA. Urinary infections in total hip arthroplasty.
Influences of prophylactic cephalosporins and catheterization. The Journal of bone and joint surgery 1976;58(8):1134-7.
21. Uckay I, Vernaz-Hegi N, Harbarth S, et al. Activity and impact on antibiotic use and costs of a dedicated infectious diseases consultant on a septic orthopaedic unit. The Journal of infection 2009;58(3):205-12.
22. Lübbeke A, Garavaglia G, Barea C, Hoffmeyer P. Why do we need hospital-based registries? The Geneva Hip Arthroplasty Registry. EFORT Publications 2010.
23. Stephan F, Sax H, Wachsmuth M, Hoffmeyer P, Clergue F, Pittet D. Reduction of urinary tract infection and antibiotic use after surgery: a controlled, prospective, before-after intervention study. Clin Infect Dis 2006;42(11):1544-51.
24. Clinical and Laboratory Standard's Institute. Standard M100-S17. 2007;Wayne PA.
25. Uckay I, Hoffmeyer P, Lew D, Pittet D. Prevention of surgical site infections in orthopaedic surgery and bone trauma: state-of-the-art update. The Journal of hospital infection;84(1):5-12.
26. Uckay I, Sax H, Gayet-Ageron A, et al. High proportion of healthcare-associated urinary tract infection in the absence of prior exposure to urinary catheter: a cross-sectional study.
Antimicrobial resistance and infection control;2(1):5.
27. Otermin I, Rivero M, Hidalgo A. [Is it necessary to delay or to put off surgery in the case of possible asymptomatic bacteriuria and orthopaedic surgery with implants?]. Enfermedades infecciosas y microbiologia clinica 2009;27(4):252-3.
28. van den Brand IC, Castelein RM. Total joint arthroplasty and incidence of postoperative bacteriuria with an indwelling catheter or intermittent catheterization with one-dose antibiotic prophylaxis: a prospective randomized trial. The Journal of arthroplasty 2001;16(7):850-5.
29. Knight RM, Pellegrini VD, Jr. Bladder management after total joint arthroplasty. The Journal of arthroplasty 1996;11(8):882-8.
30. Carpiniello VL, Cendron M, Altman HG, Malloy TR, Booth R. Treatment of urinary complications after total joint replacement in elderly females. Urology 1988;32(3):186-8.
31. Stamm WE, McKevitt M, Roberts PL, White NJ. Natural history of recurrent urinary tract infections in women. Reviews of infectious diseases 1991;13(1):77-84.
32. Tambyah PA, Maki DG. Catheter-associated urinary tract infection is rarely symptomatic:
a prospective study of 1,497 catheterized patients. Archives of internal medicine 2000;160(5):678-82.
33. Ferry SA, Holm SE, Stenlund H, Lundholm R, Monsen TJ. The natural course of uncomplicated lower urinary tract infection in women illustrated by a randomized placebo controlled study. Scandinavian journal of infectious diseases 2004;36(4):296-301.
34. Bleidorn J, Gagyor I, Kochen MM, Wegscheider K, Hummers-Pradier E. Symptomatic treatment (ibuprofen) or antibiotics (ciprofloxacin) for uncomplicated urinary tract infection?-- results of a randomized controlled pilot trial. BMC medicine;8:30.
35. Saleh K, Olson M, Resig S, et al. Predictors of wound infection in hip and knee joint replacement: results from a 20 year surveillance program. J Orthop Res 2002;20(3):506-15.
36. Uckay I, Lubbeke A, Emonet S, et al. Low incidence of haematogenous seeding to total hip and knee prostheses in patients with remote infections. The Journal of infection 2009;59(5):337-45.
37. Panagamuwa C, Glasby MJ, Peckham TJ. Dipstick screening for urinary tract infection
All patients
Leukocyturia* 181/510 (35%)
Positive nitrites* 28/510 (5%)
Positive urine culture 182/510 (36%)
Patients with positive urine culture
Covered by cefuroxime* 119/182 (65%)
-Escherichia coli 53/182 (29%)
-Klebsiella spp 4/182 (2%)
-Non-fermenting rods 2/182 (1%)
Gram-positive colonisation 100/182 (55%) -Coagulase-negative staphylococci 9/182 (5%)
-Enterococcus spp* 18/182 (10%)
Table 1 Epidemiology of urine pathologies before and three days after arthroplasty
On admission, n=510 Surgery day Day 3, n=510
Antiobiotic
99/510 (19%) 15/510 (3%) 208/510 (41%) prophylaxis
22/208 (11%) 46/208 (22%) 5/208 (2%) 14/208 (7%) 72/208 (35%) 23/208 (11%) 52/208 (25%)
*The difference between pre-and postoperative sampling is statistically significant (p ≤ 0.05)
Table 2 Group comparison stratified according to symptomatic urine tract infection
n = 510
Cystitis n = 25
Comparison p value*
No cystitis n = 485
Women 19 (76%) 293 (60%)
Age group ≤ 65 years 6 (24%) 178 (37%)
Age group 66 - 75 years 9 (36%) 163 (34%)
Age group > 75 years 10 (40%) 144 (30%)
Immune suppression+ 3 (12%) 65 (13%)
Total hip joint prosthesis 16 (64%) 274 (56%)
Total knee joint prosthesis 9 (36%) 211 (44%)
Preoperative leukocyturia 14 (56%) 0.025 165 (34%)
Preoperative positive nitrites 5 (20%) 0.001 23 (5%)
Preoperative positive culture 14 (67%) 0.041 168 (44%)
-pathogen susceptible to cefuroxime 7 (58%) 76 (40%)
Questionnaires returned 25 (100%) 421 (87%)
*Only p values ≤0.05 (two-tailed) are shown. χ²-test.
+ Diabetes mellitus, immune suppressive therapy, dialysis, cirrhosis, active cancer
Table 3 English-language literature review addressing the futility of preoperative urine analysis for the search of asymptomatic bacteriuria before arthroplasty
Author Year Population Sample
size
Study form %
bacteriuria
Treatment bacteriuria
Arthroplasty infection
Conclusion
Donovan20 1976 Hip 426 Retro- and
prospective
5% yes 1 seeding Treatment selects
Pseudomonas spp
Glynn2 1984 Hip/and knee 299 Retrospective 18% yes No seeding No postponing of surgery
necessary
Ritter12 1987 Hip and knee 364 Retrospective 13% yes No seeding Urine analysis futile
Kumar17 2008 Hip 100 Retrospective 3% yes No seeding Antibiotics are safe when
handling urine catheters
Ollivere19 2009 All orthopaedics 558 Prospective yes 4 seedings to
wounds
Treat bacteriuria
Koulouvaris7 2009 Hip and knee 19,375 Retrospective yes; 5-8 days No seeding Don’t postpone surgery
Martinez-Vélez5 2010 Hip/knee 200 Prospective - yes, for 7 days No seeding Treat bacteriuria?
Finnigan11 2012 Orthopaedic - Survey - 70% of surgeons - Teach more evidence
surgeons
Cordero-Ampuero4 2013 Hip and
hemiarthroplasty
471 Prospective 10% yes; 2-10 days No seeding No danger of bacteriuria
David6 2000 Hip and knee - Opinion paper - yes possible Treat if colony count > 10³
pathogens/ml
Rajamanickam10 2007 Hip and knee - Opinion paper yes No seeding Treat if colony count >
105* pathogens/ml
Our study 2013 Hip and knee 510 Prospective 36% no No seeding futile
