Hospital infections in 46 patients submitted to total hip replacement

Abstracts

Foram estudados 46 pacientes submetidos a artroplastia total do quadril em um Instituto de Ortopedia de São Paulo, Brasil, no período de 1993 a 1995, com o objetivo de obter a real frequência das infecções hospitalares da ferida operatória superficial e profunda que ocorrem nessa cirurgia. O estudo baseou-se no acompanhamento pré-operatório, trans-operatório e pós-operatório com seguimento mínimo de três meses de todos os pacientes, no sentido de caracterizar os agentes etiológicos das infecções e os fatores de risco que contribuem para o seu desenvolvimento. Foi observada uma freqüência total de 15.1% de infecções hospitalares, sendo 6.5% de infecção superficial da ferida operatória, 6.5% de infecção profunda e 2.1% de infecção do trato urinário. Os agentes etiológicos encontrados foram Pseudomonas aeruginosa (2 casos), Staphylococcus coagulase negativo (2 casos), Morganella morgani (1 caso) e associação de Acinetobacter calcoaceticus (2 casos). O fator de risco com significância estatística observado nesta casuística foi o tempo cirúrgico aumentado. Concluiu-se que a frequência de infecção da ferida cirúrgica superficial e profunda nas artroplastias totais de quadril foi maior, nesta casuística, do que a relatada na literatura internacional, com elevada participação de bacilos Gram-negativos como agentes etiológicos e tendo como principal fator de risco o tempo cirúrgico aumentado.

ARTIGO ORIGINAL

Hospital infections in 46 patients submitted to total hip replacement

Ana Lúcia Lei Munhoz Lima; Antonio Alci Barone

SUMMARY

We studied 46 patients submitted to total hip replacement at an Orthopaedics Institute in the city of São Paulo, Brazil, from 1993 to 1995, in order to determine the real frequency of hospital infections of the superficial and deep surgical wounds occurring in this operation. The study consisted of preoperative, transoperative and postoperative monitoring and a minimum follow-up period of three months for all patients in order to characterise the etiologic agents of the infections and risk factors contributing to the development of the latter.

The total frequency of hospital infections was 15.1%, 6.5% of them being superficial wound sepsis, 6.5% deep wound sepsis and 2.2% infection of the urinary tract. The etiologic agents detected were: Pseudomonas aeruginosa (2 cases), coagulase-negative Staphylococcus (2 cases), Morganella morgani (1 case), and association of Acinetobacter calcoaceticus (2 cases). The risk factor showing a statistically significant effect in this patient series was increased surgical time.

We conclude that the frequency of superficial and deep surgical wound sepsis in total hip replacement surgeries was higher in the present series compared to data reported in the international literature, with a high participation of Gram-negative bacilli as etiologic agents and with increased surgical time representing the major risk factor.

INTRODUCTION

The development of joint prostheses represents a great advancement in biomedical technology. The implantation of prostheses, especially of the hip and knee, is becoming increasingly frequent, with an estimated 150 thousand hip arthroplasties being performed per year in the United States and 400 thousand in the whole world. When implanted, these prostheses provide a significant reduction in the discomfort of patients with coxofemoral joint diseases and an immeasurable improvement in mobility, especially when not complicated by infectious processes.

Literature data reveal that 1 to 5% of hip prostheses become infected. Although less frequent than mechanical loss of the implant, infection is considered to be the most devastating complication since it involves prolonged hospitalisation and repeated surgical interventions possibly culminating in the definitive loss of the implant, with shortening of the limb involved, severe and permanent deformities, and even death in cases of fulminating infection. Furthermore, all the procedures involved in the attempts to resolve the infection are very expensive, with their cost being estimated at 40 to 80 million dollars/year in the United States.

With the increasing indications of the use of joint prostheses, the improvement of surgical techniques employed by specialised groups and the development of new implant models, there is an increasing necessity for more in-depth knowledge of the infections associated with these procedures.

The onset of sepsis in arthroplasty is known to be related to events occurring before, during and after surgery. The objective of the present study was to determine the frequency of hospital infections occurring in total hip arthroplasty, as well as the risk factors for their development. We defined as hospital infections all of those that occurred up to the thirtieth postoperative day. However, it should be kept in mind that the infections that develop in the operated region up to one year after surgery for prosthesis implantation may be related to the surgery itself.

All 46 patients were submitted to elective surgery, and were operated upon by the hip group of the Institute of Orthopaedics and Traumatology of the University Hospital, Faculty of Medicine, University of São Paulo, in an attempt to obtain uniform procedures and to limit the variables. Elective selection of the patients permitted us to perform a careful preoperative evaluation in which we searched for and treated pulmonary, dental, skin and urinary foci of infection, which are considerably implicated in the postoperative hematogenic dissemination of bacteria and contamination of the prosthesis.

CASES AND METHODS

A total of 46 patients were followed up from 1993 to 1995. Median patient age was 54 years(range: 22 to 80 years), 21 were males (45.7%) and 25 females (54.3%).

Preoperative evaluation: epidemiological data concerning age, sex, preexisting conditions, disease that led to the indication of arthroplasty, previous surgeries (orthopaedic or not), previous history of osteoarticular infection or infections localized elsewhere, medications being taken, smoking habit, alcohol use and abuse of injectable illicit drugs were obtained.

Clinical data were obtained by general and specific physical examination.

Laboratory tests included blood counts, blood sedimentation rate, fasting blood glucose levels, determination of serum urea and creatinine, coagulogram, urine culture and culture of nasopharyngeal secretion, with an antibiogram when bacterial growth was detected. Patients with a history of osteoarticular infection at the site to be operated upon were submitted to bone scintigraphy with technetium-labelled immunoglobulin.

The alteration detected, as well as eventual focal points of infection were treated and resolved before surgery.

Transoperative evaluation: all patients were admitted to hospital on the day of surgery. They were operated upon at the Surgical Center of the Institute of Orthopaedics and Traumatology (IOT), where the regulations for material sterilisation, room disinfection, prosthesis packing and verification of its sterilisation conditions follow the standards established by the Committee of Control of Hospital Infection of the institution. The surgical technique used in all cases was that standardised by the hip group of the IOT. The number of physicians who participated in the surgery was recorded. We also recorded whether or not cement was used for prosthesis fixation.

All patients received prophylactic antibiotic treatment with cephalothin, 2 grams intravenously during induction of anaesthesia, followed by one gram every six hours for 48 hours. Bone fragments were collected during the surgical act for culture and antibiogram and for histopathological examination.

Postoperative evaluation: all patients were assessed daily until discharge and then on the thirtieth postoperative day and at least three months later. The infections were diagnosed based on the 1988 criteria of the Center for Diseases Control (NNISS).

RESULTS

Preoperative examination: 39.1% of the patients showed no relevant disease, 15.2% had rheumatoid arthritis, 10.8% had systemic arterial hypertension, 8.6% were under corticoid treatment, 4.3% had diabetes mellitus, and 13% had a previous history of local surgery.

The diseases that led to the indication of arthroplasty were hip osteoarthrosis in 50% of cases, rheumatoid arthritis in 13%, aseptic necrosis of the femoral head in 15.2%, ankylosing spondylitis in 10.8%, sequelae of fracture of the femur in 4.3%, sequelae of hip pyoarthritis in 2.1%, epiphysiolisthesis in 2.1%, and bilateral Otto pelvis in 2.1%.

Nonspecific tests showed that 63% of the patients had an increased blood sedimentation rate (>20 mm), and 19.5% (9 patients) had a positive urine culture with more than 100 thousand colony forming units of the same bacterium, i.e. Escherichia coli in 5 cases, and Enterobacter cloacae, Enterobacter agglomerans, Providencia rettgeri and Candida sp in one patient each. Four patients (8.6%) had a Staphylococcus aureus-positive nasopharyngeal culture.

Transoperative evaluation: the surgical team ranged from three to seven doctors and surgical time ranged from 30 to 300 minutes, with a mean of 150 minutes; 54.3% of the prostheses were fixed with cement and 45.7% without cement.

In two patients there was growth of coagulase-negative Staphylococcus in the bone fragments collected during surgery. However, no signs of infectious processes were detected in any bone fragment by histopathological examination.

Postoperative evaluation: hospitalisation time ranged from six to 36 days. Surgical wound sepsis was detected in 6 patients (15.1%), three of them with superficial infection and three with deep infection. One of these patients had a concomitant urinary infection. The etiologic agents isolated from these sites are listed in Table I.

DISCUSSION

Athough the present results are better than those that could be imagined during the first years after the introduction of total hip or knee arthroplasties, the management of infections established in prosthetic joints continues to be difficult and expensive in terms of

time and resources. Thus, even though the uncertainties and difficulties continue, careful measures should be taken to prevent infection.¹ Despite the great improvement of diagnostic possibilities with the evolution of new immunological and imaging techniques, the devastating consequences of an infection, even after an early diagnosis, cannot be avoided as yet. For this reason, more in-depth investigations and the development of additional techniques are required.² Although there was a dramatic reduction in the rates of infection due to intraoperative contamination, the rates of late infection continue to be essentially the same. At present, the rate of infection throughout the useful life of the prosthesis continues to be approximately 1%.³

A survey of the literature reveals the concern with the assessment of local or distant infectious foci when joint prostheses are implanted. There are countless reports of hematogenic dissemination of these foci after prosthesis implantation, as observed for urinary foci, lower limb ulcers, pneumonia and skin abscesses, among others.^{4, 5}

Some of the results obtained with the present series showed the extreme relevance of preoperative evaluation. Asymptomatic urinary infections were detected preoperatively in 19.5% of patients (8 women and 1 man), occurring on average at 56.3 years of age, This prevalence is much higher than that observed in the general population of the same age range, as reported by Glynn and Sheehan,⁶ who found urinary infection in 4.4% of the women and 0.5% of the men.

Considering that all patients were asymptomatic, that the prevalence was higher than in the population at large, and that the possibility of hematogenic dissemination of these microorganisms exists, urine culture becomes an obligatory preoperative exam for hip arthroplasty in all cases and not only in symptomatic cases, as recommended by Charnley and Efthekar. Also preoperatively, we found 8.6% of patients with nasal Staphylooccus aureus, a prevalence similar to that reported by Hill et al.⁸ Although the correlation between carrier status without skin sisease and hematogenuc dissemination,⁸ these patients were treated with 2% mupyrocin before surgery.

There was no significant correlation between urinary infection or nasal presence of staphyococcus and development of postoperative infections.

During the transoperative period we obtained relevant results in terms of surgical time. As also reported by Fitzgerald et al.,² patients who developed postoperative infection were those submitted to a surgical time of more than 140 minutes.

The frequency of superficial and deep surgical wound infections was 8.6% at discharge and increased to 13% on the occasion of reevaluation on the thirtieth day. These data agree with those reported by Cardo et al.⁹ and Taylor et al.¹⁰ who observed an increase of about 20% in the diagnosis of surgical wound infections in patients reevaluated after discharge from hospital. In a specific study of surgical wound infections in arthroplasties of the hip and joint, Taylor et al.¹⁰ concluded that 77% of them were diagnosed after discharge from hospital. These findings show the importance of monitoring the surgical wounds after discharge, not only in order to detect their real frequency but also for early diagnosis and proper treatment, thus preventing the occurrence of deep sepsis originating from superficial sepsis with a delayed diagnosis.

We should also point out the high occurrence of Gram-negative bacilli as etiologic agents of superficial and deep post-arthroplasty infections, in contrast to the staphylococcal infections traditionally reported in the world literature.

CONCLUSIONS

The frequency of superficial and deep surgical wound infections in total hip arthroplasty was higher in the present series than the values reported in the literature.

Surgical time was the most relevant risk factors for the development of postoperative infections.

The occurrence of Gram-negative bacilli as etiologic agents of these infections was higher than the rate commonly reported in the literature for this type of surgery.

REFERENCES

Trabalho realizado pelo Grupo de Infecção do Instituto de Ortopedia e Traumatologia e pelo Departamento de Doenças Infecciosas e Parasitárias do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo.

• 1. Gillespie WJ. Prevention and management of infection atter total joint replacement. Clin Infect Dis 1997; 25: 1310-1317.
• 2. Fitzgeral Jr RH, Nolan DR, Ilstrup DM, Vanscpy RE, Washington JA, Coventry MB. Deep wound sepsis following total hip arthroplasty. J Bone Joint Surg 1977; 59a: 847-855.
• 3. Nasser S. Prevention and tratment of sepsis in total hip replacement surgery. Orthop Clin North Am 1992; 23: 265-277.
• 4. D'Ambrosia RD, Shoji H and Heater R. Secondarily infected total joint replacements by hematogenous spread. J. Bone Joint Surg. 1976; 58-a(4): 450-453.
• 5. Donovan TL, Gordon RO and Nagel DA. Urinary infections in total hip arthroplasty. J. Bone Joint Surg. 1976; 58-a(8): 1134-1137.
• 6. Glynn MK, Sheehan JM. The significance of asymptomatic bacteria in patients undergoing hip/Knee arthroplasty. Clin Orthop. 1984; 185: 151-154.
• 7. Charnley J and Eftekhar N. Postoperative infection in total prosthetic replacement arthroplasty of the hip-joint. B. J. Surg. 1969; 56(9): 641-649.
• 8. Hill, J.; Howell, A.; Blower, S.R. Effect of clothing on dispersal of Staphylococcus aureus by males and females Lancet 1974; II: 1131-1133
• 9. Cardo DM, Falk PS, Mayhall CG. Validation of Surgical Wound Surveillance. Infect. Control Hosp. Epidemiol. 1993; 14 (4): 211-215.
• 10.Taylor S, Pearce P, McKenzie M, Taylor GD. Wound infection in Total Joint Arthroplasty: Effect of Extended Wound Surveillance on Wound Infection Rates. Canadian Journal of Surgery. 1994; 37: 217-220.

Publication Dates

• Publication in this collection
  27 June 2006
• Date of issue
  Mar 2001