Microbiological Profile of Periprosthetic Joint Infections: A Retrospective Analysis from North India

ABSTRACT.

With the rise in total joint arthroplasties, prosthetic joint infections (PJIs) have become a significant complication, leading to high morbidity. The causative organisms of PJIs vary by region, and the rates of drug-resistant organisms are growing, thus complicating the initial empiric choice of antibiotics. This retrospective study analyzed records of patients with orthopedic implants and intraoperative tissue samples sent for sonication and culture at a tertiary care hospital in India. The most common organism was Staphylococcus aureus (14 out of 86 bacterial isolates, 16.3%), followed by Pseudomonas aeruginosa (12 out of 86, 13.9%), and both Staphylococcus epidermidis and Klebsiella pneumoniae (11 each out of 86, 12.8%). There was a high prevalence of multidrug-resistant organisms, and 35% of the gram-negative organisms were carbapenem resistant. Our study reveals that in our setting, PJIs are chiefly driven by multidrug resistant gram-negative bacteria.

INTRODUCTION

Periprosthetic joint infections (PJIs) are a serious complication of total joint arthroplasties (TJA), causing significant financial burden, limited mobility, and reduced quality of life. Diagnosing PJIs is challenging owing to the low sensitivity of conventional periprosthetic tissue cultures and limited availability of molecular methods such as polymerase chain reaction (PCR) and sequencing in resource-limited settings. The diagnosis and treatment of PJIs are further complicated by slow-growing organisms, biofilms, and prior antibiotic therapy.^1–3

This study aimed to identify the etiological profile of organisms from intraoperative tissue samples and sonicated implants in PJI cases to effectively guide empiric therapy.

MATERIALS AND METHODS

This retrospective study reviewed records of patients whose tissue samples and orthopedic implants were sent for sonication and culture from August 2019 to September 2022 at a tertiary care hospital in India. The data that were collected included demographic details, the affected joint, organisms isolated, and their susceptibility profile.

Prosthetic joint infection was defined using the Musculoskeletal Infection Society (MSIS) criteria.⁴ A microbiological diagnosis required the isolation of the same microorganism in two or more tissue specimens. Multidrug resistance (MDR) was defined as nonsusceptibility to at least one agent in three or more antimicrobial categories.

Data were collated using Microsoft Excel and analyzed using Stata Statistical Software: Release 17 (StataCorp LP, College Station, TX). Normality was checked with the Shapiro-Wilk test. Antimicrobial susceptibility data were exported to WHONET for analysis.

RESULTS

Implants and tissues from 270 patients were analyzed during the study period. Of these, 174 patients (64.4%) were male. The mean (SD) age was 51.9 (15.8) years. A total of 1,134 tissue specimens (mean of four per patient) and 270 explanted implants were analyzed. The affected joints were 207 (76.6%) hip joints and 63 (23.3%) knee joints.

Of the 270 patient samples, 185 (68.5%) did not yield any growth in either sonication fluid culture (SFC) or tissue culture. The 85 culture-positive implants yielded 88 isolates, with three (1.1%) showing polymicrobial growth. Among the 88 isolates, one was Candida albicans and one was Mycobacterium tuberculosis identified by GeneXpert. Tissue specimen culture positivity was 27% (73/270 patients). Three samples negative by SFC were positive by tissue culture. In 94% of cases (66/70), the same organism was isolated from both tissue and implant sonication fluid. In the 6% of discordant cases (4/70), the implant SFC isolate was considered the causative pathogen.

Of the 86 bacterial isolates, 38 (44%) were gram-positive and 48 (56%) were gram-negative. No anaerobes were isolated, consistent with previous findings from the institute.⁵ Table 1 provides an overview of the isolated organisms.

Table 1.

Overview of the isolated bacterial pathogens

Organism	No. of Isolates	%
Gram-Positive (N = 38)
Staphylococcus aureus	14	16.3
Staphylococcus epidermidis	11	12.8
Enterococcus faecalis	5	5.8
Staphylococcus haemolyticus	2	2.3
Staphylococcus hominis	4	4.7
Corynebacterium striatum	1	1.1
Staphylococcus warneri	1	1.1
Gram-Negative (N = 48)
Pseudomonas aeruginosa	12	13.9
Escherichia coli	8	9.3
Enterobacter cloacae	7	8.1
Klebsiella pneumoniae	11	12.8
Acinetobacter baumannii	3	3.5
Serratia marcescens	2	2.3
Morganella morganii	1	1.1
Salmonella Group A	1	1.1
Enterobacter hormaechei	1	1.1
Proteus mirabilis	1	1.1
Citrobacter sedlakii	1	1.1

The percentages are calculated based on the total number of bacterial isolates (N = 86).

The most common organism was Staphylococcus aureus (14/86 isolates, 16.3%), followed by Pseudomonas aeruginosa (12/86, 13.9%), and Staphylococcus epidermidis and Klebsiella pneumoniae (11/86 each, 12.8%). Half (7/14) of the S. aureus isolates were methicillin resistant.

Antibiotic susceptibility testing, performed according to Clinical and Laboratory Standards Institute recommendations, showed that the majority of gram-positive isolates were resistant to penicillin G (73.5%), erythromycin (62.9%), and gentamicin (50.0%). All gram-positive isolates were uniformly susceptible to teicoplanin and vancomycin. Among gram-negative isolates, resistance was highest to cefotaxime (86.4%), ceftazidime (85.7%), and netilmicin (66.7%). All gram-negative isolates were uniformly susceptible to colistin, tested by the microbroth dilution method. Details are shown in Figure 1 and Supplemental Figure 1. Carbapenem-resistant isolates were approximately 35%.

Figure 1.

The resistance pattern of gram-negative isolates (N = 48) against selected antibiotics. The bars represent 95% CIs. CAZ = ceftazidime; CFP = cefoperazone; COL = colistin; CTX = cefotaxime; IPM = imipenem; MEM = meropenem; NET = netilmicin; TGC = tigecycline; TZP = piperacillin tazobactam.

DISCUSSION

This retrospective study highlights the pathogen profile in culture-positive PJIs at a tertiary care center in northern India. Previous studies estimated our institutional PJI rates at 1.1%,⁵ which is considered typical for high-volume centers.⁶ High-volume centers often have better infection rates owing to experienced personnel and robust infection control practices. Our hospital serves a large population from northern India, with many patients referred for revision surgeries, making our microbial profile reflective of the region’s prevalent flora. Empiric therapy often relies on our knowledge of commonly implicated pathogens.

Periprosthetic joint infection was diagnosed following the MSIS criteria,⁴ with intraoperative cultures from periprosthetic tissue aiding diagnosis. Multiple cultures were sent, including from the implants removed. Sonication of explanted implants, which disrupts biofilms while preserving microbial viability, improved pathogen detection. Studies at our institute showed that SFC enhanced the detection of gram-positive pathogens.⁵ Sonication fluid culture, combined with tissue cultures, allowed for accurate pathogen identification.

Molecular methods such as 16S ribosomal ribonucleic acid sequencing, next-generation sequencing, and multiplex PCR assays offer varied success in PJI diagnosis.⁷ However, because of high costs, we relied on SFC and periprosthetic tissue cultures. Culture-negative PJI rates in our study were higher than those in the existing literature, likely owing to extensive prior antibiotic use, viable but uncultivable organisms, slow-growing organisms, and formation of biofilms.⁷

Our study found a predominance of gram-negative organisms, reflecting findings in the Indian subcontinent.⁵ This contrasts with studies from high-income countries, where gram-positive organisms are more common.^1,6 The rising incidence of antimicrobial resistance poses a significant challenge. In our study, the most frequently isolated pathogen was S. aureus (16.3%), followed by P. aeruginosa (13.9%), and S. epidermidis and K. pneumoniae (12.8% each). All gram-negative organisms were MDR.

Our study’s strengths include the routine practice of implant sonication, contributing to better pathogen yield, and the relevance of our microbial profile to that of other northern Indian healthcare centers. In addition, understanding the microbiological profile can help reduce the economic burden on patients by guiding appropriate empiric therapy. However, limitations include the lack of molecular methods for diagnosis, and nonuniform perioperative antibiotic protocols. Also, these data are based on a single tertiary care center, so the results may not be generalizable to other regions or countries. Finally, the lack of molecular methods in diagnosis remains a lacuna, which is especially relevant given the high rate of culture negativity. Collaboration with other centers and further studies are needed to better understand PJI diagnosis and management.

Conclusion

Periprosthetic joint infection is a significant complication of TJA, causing considerable morbidity. Our study showed a predominance of MDR gram-negative bacteria in our setting, different from data from developed nations. Timely PJI diagnosis is crucial for quality patient care, and the increasing incidence of MDR organisms underscores the need for antimicrobial stewardship. Financial constraints in resource-limited settings limit the use of molecular methods, making SFC a valuable diagnostic tool. Effective PJI management requires a multidisciplinary team, and future preventive strategies could reduce PJI incidence.

Note: Supplemental materials appear at www.ajtmh.org.