Abstract
Aims
Cirrhosis is a known comorbidity that may influence outcomes following total joint replacement (TJR). This meta-analysis evaluates the impact of cirrhosis on postoperative complications and mortality after TJR.
Methods
A systematic review and meta-analysis were conducted in accordance with PRISMA guidelines. Four databases were searched up to February 2025. Studies comparing the outcomes of TJR in cirrhotic versus non-cirrhotic patients were included. Data were analyzed using a random-effects model, and subgroup and sensitivity analyses were performed.
Results
A total of 12 studies, including 12.8 million TJR cases, were analyzed, with 111,011 cirrhotic patients. Cirrhosis was significantly associated with higher rates of mortality (odds ratio (OR) = 4.67, 95% CI 2.58 to 8.43), revision surgery (OR = 2.13, 95% CI 1.60 to 2.83), periprosthetic joint infection (OR = 2.61, 95% CI 2.11 to 3.21), implant failure (OR = 1.86, 95% CI 1.58 to 2.18), prosthetic fracture (OR = 1.89, 95% CI 1.19 to 3.01), transfusion (OR = 1.73, 95% CI 1.56 to 1.92), pneumonia (OR = 1.35, 95% CI 1.17 to 1.56), acute renal failure (OR = 2.16, 95% CI 1.35 to 3.46), readmission (OR = 1.93, 95% CI 1.70 to 2.20), and extended hospital stay (SMD = 0.49, 95% CI 0.28 to 0.69). Gastrointestinal complications (OR = 0.72) and pulmonary thromboembolism (OR = 0.25) were significantly lower. No significant differences were observed for wound complications, bleeding volume, operating time, venous thromboembolism, or cardiac events.
Conclusion
Cirrhotic patients undergoing TJR are at greater risk for mortality, complications, and healthcare resource use. These findings underscore the need for multidisciplinary preoperative evaluation and careful risk-benefit assessment to improve outcomes in this vulnerable population.
Cite this article: Bone Jt Open 2025;6(11):1394–1408.
Keywords: Cirrhosis, Hip, Knee, Arthroplasty, total hip and knee arthroplasty, liver cirrhosis, total joint replacement, periprosthetic joint infection (PJI), revision surgeries, comorbidities, pneumonia, bleeding, postoperative complications
Introduction
Arthroplasty (total joint replacement (TJR)) is one of the most highly efficacious orthopaedic procedures being performed frequently all over the world to boost the quality of life for patients with end-stage arthritis.1 With the ageing global population and a significant rise in the number of people diagnosed with arthritis, there is going to be an increased demand for this procedure.2
Therefore, it is important to understand the possible short- and long-term complications of this procedure, and the risk factors that may heighten the likelihood of these complications. Common risk factors for postoperative complications are advanced age, male sex, procedure type (bilateral, revision), arrhythmia, cerebrovascular disease, and so on.3,4
Another underlying condition posing challenges to this surgery is chronic liver disease, which is the continuous process of liver inflammation, destruction, and fibrosis.5 The combination of abnormal bone formation, osteopenia, and osteoporosis because of chronic liver disease is thought to be responsible.6,7 Cirrhosis, defined as the final stage of chronic liver disease, is also a common comorbidity among patients undergoing TJR, which puts them at risk of developing postoperative complications such as infection, prosthesis dislocation, and intravascular coagulation.8 In general, cirrhotic patients should be carefully evaluated preoperatively for any type of surgery due to their disease-specific factors, like impaired synthetic function, sarcopenia, malnutrition, coagulopathy, and portal hypertension, exposing them to higher morbidity and mortality rates postoperatively.9 Moreover, common aetiologies of cirrhosis, including hepatitis C virus (HCV), alcoholic liver disease, and nonalcoholic steatohepatitis (NASH), are independent risk factors for poor postoperative outcomes.10
Hence, a precise risk assessment of cirrhotic patients undergoing TJR as a widespread procedure seems to be essential to improve patient outcomes in this susceptible group. Recent studies regarding this topic suggested a variety of complications among these patients. Thus, to consolidate the findings of these studies, this systematic review aims to elucidate the existing data on the risk of postoperative complications of TJR in cirrhotic patients.
Methods
To ensure the accuracy and reliability of the study findings, this article adheres to the guidelines outlined in PRISMA. Additionally, the study protocol was registered on the International Prospective Register of Systematic Reviews (PROSPERO; registration ID: CRD42025641817).
Search strategy
A systematic search through four online databases, including Scopus, PubMed, Web of Science, and Embase, was conducted from their inception to 24 February 2025. The search terms included synonyms and related terms to cirrhosis ("cirrhosis" OR "Primary Biliary Cholangitis" OR "Secondary Biliary Cholangitis" OR "hepatic fibrosis") and arthroplasty ("joint arthroplasty" OR "total hip arthroplasty" OR "total knee arthroplasty") along with relevant keywords using boolean operators. Furthermore, the reference lists of the included papers were manually checked for any relevant studies. The comprehensive search strategy employed to query each of the databases is outlined in the Supplementary Material.
Eligibility criteria and study selection
Studies investigating the impact of liver cirrhosis on total knee or hip arthroplasty outcomes compared to control cases (patients without cirrhosis) were included in this meta-analysis. Non-original articles, case reports, case series, and editorial letters were excluded. In addition, studies focusing on arthroplasty of joints other than the hip or knee were excluded.
In a two-phase process, two authors (A-MA, M-TP-F) independently conducted title and abstract screening, followed by independent screening of the full texts to determine the final included studies. Any disagreements were resolved by consultation with a senior third author (AK), whose decision was considered final.
A comprehensive literature search was performed across four databases: PubMed, Embase, Web of Science, and Scopus. After the removal of duplicate records, 287 studies remained. Of these, 254 studies were excluded during the title and abstract screening phase. The remaining 33 studies were assessed for full-text retrieval, all of which were accessible. Full-text screening resulted in the exclusion of 21 studies: five due to a lack of focus on TJR patients, five due to irrelevance to cirrhosis, five due to the absence of at least one control group, three due to incomplete demographic data, and three due to being published in languages other than English. In total, 12 studies met the inclusion criteria and were retained for the systematic review.8,10-20 The study selection process is detailed in the PRISMA flow diagram (Figure 1).
Fig. 1.
PRISMA chart. TJR, total joint replacement.
Data extraction and quality assessment
The bibliographic details including study title, primary author’s name, publication year, study design, the number of cases and control cohorts, demographic data of patients, type of arthroplasty, and reported outcomes of interest were compiled into a spreadsheet by two independent authors (A-MA, FS).
The primary outcomes of interest were mortality and revision surgery. Secondary outcomes were: anaemia, acute renal failure (ARF), cardiac complications, gastrointestinal (GI) complications, implant failure, mechanical complications, periprosthetic joint infection (PJI), pneumonia, prosthetic fracture, pulmonary thromboembolism (PTE), readmission, transfusion, venous thromboembolism (VTE), wound complications, bleeding volume, hospital charges, length of stay, and operating time.
The quality assessment of the included studies was conducted using the Newcastle-Ottawa Scale (NOS)21 by two authors (HMS, FS) independently.
Statistical analysis
Data analysis was performed using a random-effects meta-analysis model. For dichotomous outcomes, odds ratios (ORs) were calculated and reported, while continuous outcomes were assessed using standardized mean differences (SMDs). Publication bias was evaluated through Harbord’s test and funnel plots. Heterogeneity was quantified using the I² statistic. Sensitivity analyses were conducted to identify potential outlier studies.
Subgroup analysis based on the type of TJR was performed to determine the effect size in each TJR subgroup and to determine between-group differences. Three subgroups were defined as follows: total knee arthroplasty (TKA; for studies that included patients undergoing TKA), total hip arthroplasty (THA; for studies that included patients undergoing THA), and TJR (for studies that did not separate data on TKA and THA patients and reported the outcomes cumulatively).
All analyses were carried out using R software version 4.4.2 with the meta package (R Foundation for Statistical Computing, Austria). A p-value of less than 0.05 was considered statistically significant.
Results
A total of 12,810,080 arthroplasty cases (51.5% male) were included in this study, of which 111,011 (0.87%) were cirrhotic. All of the 12 studies included in the meta-analysis were retrospective cohort studies. The studies were conducted in the USA, Taiwan, South Korea, and Denmark. A detailed overview of the baseline characteristics of the included studies is presented in Table I.
Table I.
Baseline characteristics of the included studies.
| Study | Country | Study design | Study period | Procedure | Database used | Sample size, N | Number of included cirrhotics, n (%) | Number of included controls, n (%) | Mean age in cirrhotics, yrs (SD) | Mean age in control group, yrs (SD) | Mean age in total population, yrs (SD) | Male/female in total population | Mean follow-up (SD) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Sequeira et al 20218 | USA | Retrospective database review cohort | 2006 to 2013 | THA | PearlDiver | 741,078 | 18,321 (2.47) | 722,757 (97.53) | 69.77 (6.61) | 72.18 (6.16) | 72.12 | 298,188/442,890 1:0.67 | 1 yr |
| Shih et al 200411 | Taiwan | Retrospective cohort | 1992 to 2000 | THA | - | 84 | 42 (50) | 42 (50) | 67 (7) | 67 (7) | 67 (7) | 19/32 1:168 | 42 mths (8 to 128) complete follow-up 41 mths (26 to 128) survivors |
| Cheppalli et al 202320 | USA | Retrospective database review cohort | 2016 to 2019 | THA | NIS | 367,894 | 1134 (0.3) | 366,760 (99.7) | 64.6 (10.0) | 65.9 (11.39) | 65.9 | 205,742/165,152 1:0.80 | N/A |
| Bell et al 202110 | USA | Retrospective database review cohort | 2006 to 2013 | TKA | PearlDiver | 1,734,568 | 18,129 (1.05) | 1,716,439 (98.95) | 67.88 (6.40) | 71.70 (5.96) | 71.66 | 662,886/1,071,682 1:1.61 | At least 1 yr |
| Metikala et al 202319 | USA | Retrospective database review cohort | 2016 to 2019 | TKA | NIS | 558,256 | 1,670 (0.3) | 556,586 (99.70) | 65.91 (5.46) | 66.72 (5.53) | 66.71 | 214,836/343,420 1:1.60 | N/A |
| Cohen et al 200518 | USA | Retrospective cohort | 1986 to 2002 | THA/TKA | - | 122 | 29 (23.8) | 93 (76.2) | 66.7 (10.7) | 68.2 (14.1) | 67.84 | 31/91 1:2.94 | 30 days |
| Seol et al 201717 | South Korea | Retrospective cohort | 2007 to 2015 | THA/TKA | - | 220 | 110 (50) | 110 (50) | 68.9 (10.9) | 72.4 (12.3) | 70.65 | 139/81 1:0.58 | N/A |
| Newman et al 201616 | USA | Retrospective database review cohort | 2000 to 2011 | THA/TKA | NIS | 8,412,701 | 68,865 (0.81) | 8,343,836 (99.19) | 66.3 | 66.4 | 66.4 | 4,827,847/3,584,854 1:0.74 | N/A |
| Kuo et al 201615 | Taiwan | Retrospective database review cohort | 2001 to 2010 | TKA | NHIRD (LHID) | 4,619 | 124 (2.68) (HBV+) | 4,495 (97.32) | 67.9 (HBV+) | 70.13 | 70.07 | 1,184/3,435 1:2.90 | N/A |
| Deleuran et al 201514 | Denmark | Retrospective database review cohort | 1995 to 2011 | THA/TKA | NPR | 109,522 | 363 (0.33) | 109,159 (99.67) | 66 (median) | 69 (median) | N/A | 44,951/64,571 1:1.43 | 1 yr |
| Jiang et al 201413 | USA | Retrospective database review cohort | 2000 to 2010 from NIS, variable periods from SID | THA/TKA | NIS - SID | 880,786 | 2,109 (0.24) | 878,677 (99.76) | 62.3 (11.1) in THA (p < 0.05) 64.3 (9.6) in TKA (p < 0.05) 63.5 | 66.2 (12.6) in THA, 67.5 (10.4) in TKA 67.0 | 67.0 | 343,081/537,705 1:1.57 | At least 6 mths |
| Tiberi et al 201412 | USA | Retrospective case-control | 2000 to 2012 | THA/TKA | (Institutional database) | 230 | 115 (50) | 115 (50) | 62.9 | ≈ 62.9 (Matched control) | 62.5 | 118/112 1:0.95 | 4.4 yrs (3 mths to 12.1 yrs) |
HBV, hepatitis B virus; LHID, Longitudinal Health Insurance Database; N/A, not available; NHIRD, National Health Insurance Research Database; NIS, Nationwide Inpatient Sample; NPR, National Patient Registry ; SID, State Inpatient Database; THA, total hip arthroplasty; TKA, total knee arthroplasty.
Study quality was assessed using the Newcastle-Ottawa Scale (NOS).21 Six studies demonstrated high methodological quality (NOS ≥ 7), while the other six were rated as moderate quality, primarily due to limitations in follow-up duration and adequacy of follow-up (Table II).
Table II.
Quality assessment of the included studies.
| Authors | Year | Selection | Comparability | Outcome | Total Score | QA | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Representativeness of the exposed cohort (0/1) | Selection of the non-exposed cohort (0/1) | Ascertainment of exposure (0/1) | Outcome of interest not present at start (0/1) | Study controls for main factor (0/1) | Study controls for additional factors (0/1) | Assessment of outcome (0/1) | Was follow-up long enough for outcomes to occur (0/1) | Adequacy of follow-up/completeness (0/1) | ||||
| Sequeira | 2021 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 8 | High |
| Shih | 2004 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 8 | High |
| Cheppalli | 2023 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 0 | 0 | 6 | Mod |
| Bell | 2021 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 8 | High |
| Metikala | 2023 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 0 | 0 | 6 | Mod |
| Cohen | 2005 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 9 | High |
| Seol | 2017 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 7 | Mod |
| Newman | 2016 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 0 | 0 | 6 | Mod |
| Kuo | 2016 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 0 | 0 | 6 | Mod |
| Deleuran | 2015 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 8 | High |
| Jiang | 2014 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 0 | 0 | 6 | Mod |
| Tiberi | 2014 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 9 | High |
Mod, moderate; QA, quality assessment.
Meta-analyses
Primary outcomes: Our meta-analysis demonstrated a significantly higher rate of mortality (OR = 4.67, 95% CI 2.58 to 8.43, I² = 57.9%, p < 0.001) (Figure 2) and revision surgery (OR = 2.13, 95% CI 1.60 to 2.83, I² = 91.8%, p < 0.001) in cirrhotic patients compared to controls (Figure 3). This result remained consistent for mortality after removing each study through the sensitivity analysis. However, the revision rate was comparable between cirrhotic and control cases following the exclusion of the Bell et al10 study (Supplementary Material).
Fig. 2.
Forest plot of mortality complications comparison between cirrhosis and non-cirrhosis groups undergoing total joint replacement (TJR) (odds ratio (OR) with 95% CI). THA, total hip arthroplasty; TKA, total knee arthroplasty.
Fig. 3.
Forest plot of revision comparison between cirrhosis and non-cirrhosis groups undergoing total joint replacement (TJR) (odds ratio (OR) with 95% CI). THA, total hip arthroplasty; TKA, total knee arthroplasty.
Secondary outcomes
According to surgical outcomes, our study illustrated a significantly higher rate of prosthetic fracture (OR = 1.89, 95% CI 1.19 to 3.01, I² = 81.8%, p = 0.007) (Figure 4) and periprosthetic joint infection (OR = 2.61, 95% CI 2.11 to 3.21, I² = 83.6%, p < 0.001) (Figure 5) in cirrhotic cases compared to non-cirrhotic cases undergoing TJR. In addition, cirrhotic patients demonstrated a considerably higher rate of implant failure (OR = 1.86, 95% CI 1.58 to 2.18, I² = 58.5%, p < 0.001) and mechanical loosening (OR = 2.07, 95% CI 1.16 to 3.70, I² = 0.0%, p = 0.014) (Supplementary Material).
Fig. 4.
Forest plot of prosthetic fracture comparison between cirrhosis and non-cirrhosis groups undergoing total joint replacement (odds ratio (OR) with 95% CI). THA, total hip arthroplasty; TKA, total knee arthroplasty.
Fig. 5.
Forest plot of periprosthetic joint infection comparison between cirrhosis and non-cirrhosis groups undergoing total joint replacement (TJR) (odds ratio (OR) with 95% CI). THA, total hip arthroplasty; TKA, total knee arthroplasty.
However, there was no significant difference between the two groups of patients regarding wound complications (OR = 1.76, 95% CI 0.91 to 3.41, I² = 62.2%, p = 0.093), operating time (SMD = 0.5830, 95% CI –0.0629 to 1.2289, I² = 89.5%, p = 0.077) in minutes, and bleeding volume (SMD = 0.6364, 95% CI –0.0325 to 1.3053, I² = 89.9%, p = 0.062) in ml (Supplementary Material).
By performing leave-one-out analyses, all outcome results remained robust except for wound complications, which were significantly higher in cirrhotic patients after the removal of the Metikala et al19 study. In addition, bleeding volume and operating time were considerably higher in the cirrhosis group after the removal of the Seol et al17 study (Supplementary Material).
This investigation revealed that cirrhotic patients show a more extended length of stay (SMD = 0.4878, 95% CI 0.2819 to 0.6938, I² = 99.5%, p < 0.001) following TJR (Figure 6), as well as a significantly higher rate of readmission (OR = 1.93, 95% CI 1.70 to 2.20, I² = 90.6%, p < 0.001) and considerably higher hospital charges (SMD = 0.3525, 95% CI 0.1985 to 0.5065, I² = 99.4%, p < 0.001) (Supplementary Material).
Fig. 6.
Forest plot of length of stay comparison between cirrhosis and non-cirrhosis groups undergoing total joint replacement (TJR) (odds ratio (OR) with 95% CI). SMD, standardized mean difference; THA, total hip arthroplasty; TKA, total knee arthroplasty.
Conducting a sensitivity analysis showed that by removing the Seol et al17 study hospital charges were significantly higher in cirrhotic patients. However, the results of readmission rate and length of stay remained unchanged after excluding each study one by one (Supplementary Material).
Medical complications
The cirrhosis group showed a significantly higher rate of transfusion (OR = 1.73, 95% CI 1.56 to 1.92, I² = 71.6%, p < 0.001) (Figure 7) and anaemia (OR = 1.10, 95% CI 1.04 to 1.17, I² = 83.9%, p = 0.002). However, PTE (OR = 0.25, 95% CI 0.07 to 0.93, I² = 96.0%, p = 0.039) incidence was less prevalent among cirrhotic cases, and VTE incidence (OR = 0.97, 95% CI 0.87 to 1.08, I² = 0.0%, p = 0.588) was comparable between the two groups of patients (Supplementary Material).
Fig. 7.
Forest plot of transfusion comparison between cirrhosis and non-cirrhosis groups undergoing total joint replacement (TJR) (odds ratio (OR) with 95% CI). THA, total hip arthroplasty; TKA, total knee arthroplasty.
However, sensitivity analysis showed that by removing Cheppalli et al20 PTE rate was not considerably different between the two groups. Furthermore, our leave-one-out analysis demonstrated that by excluding the THA subgroup of the Newman et al16 study, anaemia was no longer higher in cirrhotic patients. The results of transfusion and VTE were robust following sensitivity analysis (Supplementary Material).
While pneumonia (OR = 1.35, 95% CI 1.17 to 1.56, I² = 41.2%, p < 0.001) was considerably higher in cirrhotic patients compared to controls (Figure 8), there was no significant difference between cirrhosis and the control group in terms of cardiac complications (OR = 0.42, 95% CI 0.04 to 3.93, I² = 51.8%, p = 0.445) (Supplementary Material).
Fig. 8.
Forest plot of pneumonia comparison between cirrhosis and non-cirrhosis groups undergoing total joint replacement (odds ratio (OR) with 95% CI). THA, total hip arthroplasty; TKA, total knee arthroplasty.
Sensitivity analysis showed a lower rate of cardiac complications in cirrhotic patients after removing the Shih et al11 study. Our findings on pneumonia remained consistent after sensitivity analysis.
While ARF (OR = 2.16, 95% CI 1.35 to 3.46, I² = 98.2%, p = 0.001) was significantly more incident in cirrhotic patients, there was a considerably lower rate of gastrointestinal (GI) complications (OR = 0.72, 95% CI 0.58 to 0.90, I² = 47.6%, p = 0.004) in the cirrhosis group (Supplementary Material).
According to our sensitivity analysis, GI complications were comparable between the two groups after removing the Newman et al16 study. The ARF result was robust following leave-one-out analysis (Supplementary Material).
A summary of the results of all the outcomes investigated in this meta-analysis is illustrated in Table III.
Table III.
Summary of meta-analysis results.
| Variable | Number of studies | Number observations | Effect size (OR/SMD (95% CI)) |
|---|---|---|---|
| Anaemia | 2 | 8,415,999 | 1.101 (1.0372 to 1.1687) |
| ARF | 4 | 8,418,461 | 2.1624 (1.3524 to 3.4575) |
| Cardiac complications | 3 | 5,614 | 0.4171 (0.0443 to 3.9295) |
| GI complications | 3 | 8,413,015 | 0.7233 (0.5819 to 0.899) |
| Implant failure | 3 | 2,475,730 | 1.8568 (1.583 to 2.178) |
| Mechanical complications | 2 | 5,530 | 2.0706 (1.1599 to 3.6963) |
| Mortality | 7 | 996,274 | 4.6688 (2.5844 to 8.4343) |
| PJI | 7 | 3,476,103 | 2.6052 (2.1117 to 3.214) |
| Pneumonia | 3 | 8,418,231 | 1.3544 (1.1729 to 1.564) |
| Prosthetic fracture | 5 | 2,481,260 | 1.8912 (1.1898 to 3.006) |
| PTE | 3 | 8,418,231 | 0.2498 (0.0668 to 0.9342) |
| Readmission | 5 | 3,466,184 | 1.9331 (1.6951 to 2.2044) |
| Revision surgery | 3 | 2,475,876 | 2.1293 (1.6049 to 2.8251) |
| Transfusion | 6 | 3,362,192 | 1.7298 (1.5591 to 1.9192) |
| VTE | 3 | 8,418,231 | 0.97 (0.8687 to 1.0831) |
| Wound complications | 4 | 8,418,315 | 1.7627 (0.9103 to 3.4136) |
| Bleeding volume | 3 | 444 | 0.6364 (-0.0325 to 1.3053) |
| Hospital charges | 4 | 3,358,664 | 0.3525 (0.1985 to 0.5065) |
| Length of stay | 8 | 3,359,338 | 0.4878 (0.2819 to 0.6938) |
| Operating time | 3 | 444 | 0.583 (-0.0629 to 1.2289) |
ARF, acute renal failure; GI, gastrointestinal; OR, odds ratio; PJI, periprosthetic joint infection; PTE, pulmonary thromboembolism; SMD, standardized mean difference; VTE, venous thromboembolism.
Subgroup analysis (based on type of TJR)
Our subgroup analysis based on the type of TJR demonstrated o significant between-group difference, except for revision surgery (p < 0.001), prosthetic fracture (p-value < 0.001), implant failure (p = 0.049), and transfusion (p = 0.028), all indicating a higher rate of complications in the TKA subgroup compared to THA. The Supplementary Material comprehensively depicts our subgroup analysis results.
Publication bias
Regarding publication bias, we investigated the funnel asymmetry for mortality, PJI, transfusion, and readmission (outcomes with the highest number of included studies), and none of them showed significant asymmetry and publication bias (p-values of 0.450, 0.820, 0.730, 0.110, respectively).
Discussion
By combining the data from 12 studies, this review demonstrated that cirrhotic patients experienced significantly higher rates of mortality and postoperative complications following TJR. Moreover, cirrhosis was associated with more extended hospital stays and more readmissions post-TJR. Previous studies have acknowledged the importance of cirrhosis in the prognosis of surgical procedures; however, to the best of our knowledge, this meta-analysis represents the most comprehensive examination of the impact of cirrhosis on THA and TKA outcomes.
In our meta-analysis, cirrhosis was associated with a higher risk of mortality after both THA and TKA. Similar findings were also reported in previous studies. A meta-analysis of 14 articles by Kohn et al22 demonstrated that the odds of mortality in HCV-infected patients were 9.37 times higher than in uninfected patients. Another report of 51 patients with cirrhosis who had undergone TKA also indicated an increased likelihood of mortality in cirrhotic patients.11 These findings, however, somehow contrast with another study that reported no significant difference in 90-day mortality between patients with moderate to severe cirrhosis and those with mild cirrhosis undergoing TKA,23 despite the higher mortality rate in the former group. It is worth noting that the population of this study was small, consisting of only 59 patients. Overall, these findings support considering cirrhosis as a predictor for poorer prognosis of TJR patients alongside other known risk factors.24,25
Our analysis revealed that cirrhotic patients undergoing TJR had a 76% higher risk of postoperative wound complications. Specifically, cirrhosis was associated with a nearly three times higher rate of PJI, twice the risk of pneumonia, and 2.1 times more revision surgeries, regardless of the operated joint. These findings align with a previous meta-analysis demonstrating that the odds of PJI are 1.98 times higher in HCV-infected patients, even after adjusting for potential confounders.22 Another meta-analysis also identified cirrhosis as a non-modifiable risk factor for PJI after THA.26 Similar findings have been reported in non-randomized studies.11,13,27 The largest of these studies, which compared postoperative complications in 1,734,568 patients after TKA, demonstrated a 79% increase in the risk of PJI among cirrhotic patients.10 Additionally, the studies demonstrated a higher risk of pneumonia in cirrhotic patients post-TJR.16,28 Given that many revision surgeries are due to PJI, it is logical that the risk of revision is also increased in cirrhotic patients.29 The higher susceptibility to infection in cirrhotic patients is likely due to cirrhosis-associated reticule-endothelial system impairment, defective phagocytic activity of neutrophils and macrophages, gut dysbiosis, and increased bacterial dislocation.23,30
Moreover, this study found that cirrhotic patients had a 2.25-fold higher rate of transfusion. This finding is consistent with data from 367,894 patients in the Nationwide Inpatient Sample (NIS) database who underwent THA and showed a threefold higher risk of transfusion in cirrhotic patients.31 This increased risk may be due to a higher risk of postoperative bleeding and acute renal failure in these patients.17 Additionally, cirrhosis was associated with a 3.5-times higher risk of wound complications, a 2.3-times higher risk of prosthetic fracture, a 1.6-times higher risk of mechanical complications, and a 1.9-times higher risk of implantation failure in this study, irrespective of the surgical joint. Poor nutrition, steroid dependence, hypogonadism, hypoalbuminemia, and demineralization due to hepatic osteodystrophy are considered the main factors for these complications.32-34 However, the underlying pathophysiology of these complications still needs to be studied. It is noteworthy that our meta-analysis unveiled a significantly lower rate of PTE in cirrhotic patients, aligning with Barba et al’s35 study, which found that PTE was four times higher in patients without liver disease compared to those with moderate to severe liver disease (1.6% vs 0.4%). Similarly, Cheppalli et al20 reported that PTE incidence was significantly lower in cirrhotic cases undergoing THA. That said, some studies reported an increased rate of PTE in cirrhosis patients, showing conflicting results on this matter.36,37 The observed decrease in PTE rate in cirrhotic patients compared to the control group in our study could be explained by the probable development of coagulopathy as a consequence of cirrhosis. As mentioned in a previous systematic review,38 a decreased level of anticoagulant factor is expected in patients with chronic liver disease; however, as the disease progresses, a reduction in coagulation factors produced by hepatocytes occurs. This suggests that clinicians should anticipate a spectrum of clotting to bleeding complications, emphasizing the importance of personalized decision-making in managing these patients. Future studies should investigate the role of cirrhosis aetiology, severity, and follow-up duration as potential influencing factors on the comparative risk of PTE in cirrhosis cases.
Our study also demonstrated a considerably lower rate of gastrointestinal complications in the cirrhotic group. However, sensitivity analysis revealed that after excluding the Newman et al16 study, this difference was no longer significant. This could be explained by the difference in the studies’ methodology, as in Tiberi et al12 and Shih et al11 studies cirrhotic and non-cirrhotic cases were matched for demographic and baseline characteristics, whereas in the Newman et al study,16 the two groups of patients had a significant difference in terms of demographic variables with considerably higher age in the non-cirrhotic group, which may explain to some extent the higher GI complication in this subpopulation. These findings, along with the heterogeneity between the studies, necessitate cautious interpretation of our results regarding GI complications. Further research is needed to confirm these findings and address the inconsistencies across studies.
This study also demonstrated that the length of stay in cirrhotic patients was 0.49 days longer after TJR. This finding is comparable to the outcome observed in the NIS database, which included 8,412,701 TJR patients. The data showed that patients undergoing TKA and THA had lengths of stay that were 0.30 and 0.48 days longer, respectively.16 This is important, given that approximately one million TJR surgeries are performed yearly in the USA.39 Furthermore, the annual rate of THA and TKA procedures is projected to increase to 635,000 and 1.26 million by 2030, respectively.2 Moreover, the increased length of stay associated with cirrhosis contributes to higher healthcare costs.
It is noteworthy that different clinical factors, such as cirrhosis severity, aetiology, and existing comorbidities, could have affected the outcomes of TJR in cirrhotic cases. Shih et al11 reported that hepatic decompensation history or variceal bleeding was an independent predictor for the incidence of complications in patients undergoing TKA. Regarding the prognostic role of Child-Pugh classification in cirrhotic cases,40 the Cohen et al18 study reported no significant association between this classification and major complications. Additionally, Seol et al’s study17 reported an insignificant association between the Child-Pugh classification and surgical complications. It should be noted that the small sample size of these studies may have contributed to the insignificant difference between the Child-Pugh classes. This highlights the importance of conducting future large-population studies to further elucidate the role of cirrhosis severity, based on different classification systems, as a prognostic factor in TJR patients. Furthermore, the Tiberi et al12 study proposed the Model for End-stage Liver Disease (MELD) as a novel scoring system to assess the prognosis of cirrhosis cases undergoing TJR for use by orthopaedic surgeons. This scoring system, compared to the Child-Pugh system, is less complex for use by orthopaedic surgeons with limited experience in assessing the clinical features of cirrhosis, such as encephalopathy and ascites. This study showed that MELD > 10 was associated with a three-times higher risk of overall complications and four-times higher mortality.
Regarding aetiological subtypes of cirrhosis, Newman et al16 stratified cirrhosis aetiology as alcoholic, viral, and other causes, and reported that, compared to controls, all three subpopulations demonstrated a significantly higher complication rate. In addition, the alcoholic subpopulation showed the highest effect size, with 2.51- and 2.25-times higher odds of complications in TKA and THA cases, respectively. Sequeira et al’s8 study, focusing on THA patients, demonstrated that both 90-day and one-year complications were significantly higher in alcoholic and viral cirrhosis patients, in comparison to controls, while the other causes group effect was more pronounced in terms of 90-day medical complications. A similar trend was observed in Bell et al’s10 study focusing on TKA cases, that while viral and alcoholic subgroups showed significant effect in both 90-day and one-year medical complications, the effect of cirrhosis caused by other aetiologies was more considerable in terms of 90-day medical complications. Taken all together, while there is a trend towards a higher complication rate in alcoholic and viral subpopulations, there is a need for future studies focusing on the role of aetiology in the complication rate following arthroplasty in both TKA and THA patients.
Some limitations should be considered in this study. First, although some studies investigated the cirrhosis effect on TJR outcomes using matched control and diseased groups, most of the included studies did not match the exposure and control groups for potential confounders, which highlights the need for effect-size adjustment for demographic and clinical confounders in future studies. Second, most of the included studies were observational, which may have introduced selection bias and unaccounted confounding variables, potentially impacting the observed outcomes. Third, significant heterogeneity was observed among the studies in some of our evaluations, which was not solely explained by differences in the type of joint arthroplasty. Previous evaluations demonstrated that patients with alcohol-related cirrhosis experienced more complications compared to other subtypes of cirrhosis.16 However, this meta-analysis was based on study-level data, limiting our ability to perform in-depth subgroup analyses based on the cause of cirrhosis or other factors like surgical techniques, types of prostheses, and perioperative management strategies. Fourth, the evaluations mainly compared short-term outcomes between cirrhotic and non-cirrhotic patients, and the long-term outcomes of cirrhotic TJR candidates remain understudied. Fifth, where a probable association between cirrhosis severity and aetiology, as well as existing comorbidities, on the complication rate following TJR is expected, according to insufficient data in the relevant papers, we could not conduct a subgroup analysis to perform a quantitative exploration of this association. More research is needed to better understand the long-term complications of cirrhotic patients post-TJR.
In conclusion, this study indicates that cirrhotic patients experience higher risks of mortality and complications, as well as longer hospital stays and more frequent readmissions following TJR. These findings should not discourage the use of TJR in cirrhotic patients. Instead, they highlight the importance of careful decision-making by surgeons, who must weigh the potential postoperative outcomes and complications against the benefits based on individualized patient factors. Additionally, a multidisciplinary perioperative approach is essential to minimize the increased complications associated with cirrhosis.
Take home message
- Cirrhotic patients undergoing total joint replacement are at greater risk for mortality, complications, and healthcare resource use.
- These findings underscore the need for multidisciplinary preoperative evaluation and careful risk-benefit assessment to improve outcomes in this vulnerable population.
Author contributions
A. Asgari: Conceptualization, Data curation, Investigation, Methodology, Project administration, Validation, Writing – original draft
F. Shaker: Conceptualization, Data curation, Investigation, Methodology, Project administration, Validation, Writing – original draft
M. Malekshahi: Conceptualization, Data curation, Investigation, Writing – original draft
K. Tavakoli: Investigation, Methodology, Writing – original draft
F. Behroo: Data curation, Investigation, Writing – original draft
M. Pahlevan-Fallahy: Conceptualization, Formal analysis, Methodology, Software, Validation
A. Kasaeian: Project administration, Supervision, Validation, Writing – review & editing
Funding statement
The authors received no financial support for the research, authorship, and/or publication of this article.
Data sharing
All data generated or analyzed during this study are included in the published article and/or in the supplementary material.
Ethical review statement
Informed consent was waived due to the nature of the study.
Trial registration number
The protocol of this study was registered in the PROSPERO register (CRD42025641817).
Supplementary material
Search strategy of each database, and subgroup analysis (based on total joint replacement type) results, and figures showing the sensitivity analysis for the complications.
© 2025 Asgari et al. This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives (CC BY-NC-ND 4.0) licence, which permits the copying and redistribution of the work only, and provided the original author and source are credited. See https://creativecommons.org/licenses/by-nc-nd/4.0/
Data Availability
All data generated or analyzed during this study are included in the published article and/or in the supplementary material.
References
- 1. Kamaruzaman H, Kinghorn P, Oppong R. Cost-effectiveness of surgical interventions for the management of osteoarthritis: a systematic review of the literature. BMC Musculoskelet Disord. 2017;18(1):183. doi: 10.1186/s12891-017-1540-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Sloan M, Premkumar A, Sheth NP. Projected volume of primary total joint arthroplasty in the U.S., 2014 to 2030. J Bone Joint Surg Am. 2014;100-A(17):1455–1460. doi: 10.2106/JBJS.17.01617. [DOI] [PubMed] [Google Scholar]
- 3. Ko MS, Choi CH, Yoon HK, et al. Risk factors of postoperative complications following total knee arthroplasty in Korea: a nationwide retrospective cohort study. Medicine (Baltimore) 2021;100(48):e28052. doi: 10.1097/MD.0000000000028052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Ma L-L, Yu X-R, Weng X-S, et al. Possible risk factors for severe complications occurring after primary total knee arthroplasty. Chin Med Sci J. 2022;37(4):303–308. doi: 10.24920/003938. [DOI] [PubMed] [Google Scholar]
- 5. Tapper EB, Goldberg D, Parikh ND, et al. The liver cirrhosis network cohort study: cirrhosis definition, study population, and endpoints. Am J Gastroenterol. 2025;120(3):570–575. doi: 10.14309/ajg.0000000000002953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Thien TM, Chatziagorou G, Garellick G, et al. Periprosthetic femoral fracture within two years after total hip replacement: analysis of 437,629 operations in the nordic arthroplasty register association database. J Bone Joint Surg Am. 2014;96-A(19):e167. doi: 10.2106/JBJS.M.00643. [DOI] [PubMed] [Google Scholar]
- 7. Thomson M, Scott A, Trost S, Lake J, Lim N. Low screening rates and high prevalence of osteoporosis in cirrhosis: a real-world retrospective analysis. Aliment Pharmacol Ther. 2024;59(4):535–546. doi: 10.1111/apt.17823. [DOI] [PubMed] [Google Scholar]
- 8. Sequeira SB, Labaran LA, Bell JE, Amin RM, Rao SS, Werner BC. Compensated cirrhosis is associated with increased risk of complications following total hip arthroplasty in a large medicare database. J Arthroplasty. 2021;36(4):1361–1366. doi: 10.1016/j.arth.2020.10.008. [DOI] [PubMed] [Google Scholar]
- 9. Kaltenbach MG, Mahmud N. Assessing the risk of surgery in patients with cirrhosis. Hepatol Commun. 2023;7(4):e0086. doi: 10.1097/HC9.0000000000000086. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Bell JE, Amin R, Labaran LA, Sequeira SB, Rao SS, Werner BC. Impact of compensated cirrhosis etiology on postoperative outcomes following total knee arthroplasty. J Arthroplasty. 2021;36(1):148–153. doi: 10.1016/j.arth.2020.07.019. [DOI] [PubMed] [Google Scholar]
- 11. Shih L-Y, Cheng C-Y, Chang C-H, Hsu K-Y, Hsu RW-W, Shih H-N. Total knee arthroplasty in patients with liver cirrhosis. J Bone Joint Surg Am. 2004;86-A(2):335–341. doi: 10.2106/00004623-200402000-00017. [DOI] [PubMed] [Google Scholar]
- 12. Tiberi JV, Hansen V, El-Abbadi N, Bedair H. Increased complication rates after hip and knee arthroplasty in patients with cirrhosis of the liver. Clin Orthop Relat Res. 2014;472(9):2774–2778. doi: 10.1007/s11999-014-3681-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Jiang SL, Schairer WW, Bozic KJ. Increased rates of periprosthetic joint infection in patients with cirrhosis undergoing total joint arthroplasty. Clin Orthop Relat Res. 2014;472(8):2483–2491. doi: 10.1007/s11999-014-3593-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Deleuran T, Vilstrup H, Overgaard S, Jepsen P. Cirrhosis patients have increased risk of complications after hip or knee arthroplasty. Acta Orthop. 2015;86(1):108–113. doi: 10.3109/17453674.2014.961397. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Kuo S-J, Huang P-H, Chang C-C, et al. Hepatitis B virus infection is a risk factor for periprosthetic joint infection among males after total knee arthroplasty: a Taiwanese nationwide population-based study. Medicine (Baltimore) 2016;95(22):e3806. doi: 10.1097/MD.0000000000003806. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Newman JM, Schiltz NK, Mudd CD, Szubski CR, Klika AK, Barsoum WK. Impact of cirrhosis on resource use and inpatient complications in patients undergoing total knee and hip arthroplasty. J Arthroplasty. 2016;31(11):2395–2401. doi: 10.1016/j.arth.2016.04.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Seol YJ, Yoon TR, Lee DH, Lee SH, Park KS. Outcome analysis of hip or knee arthroplasty in patients with cirrhotic liver disease. J Orthop. 2017;14(1):171–175. doi: 10.1016/j.jor.2016.12.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Cohen SM, Te HS, Levitsky J. Operative risk of total hip and knee arthroplasty in cirrhotic patients. J Arthroplasty. 2005;20(4):460–466. doi: 10.1016/j.arth.2004.05.004. [DOI] [PubMed] [Google Scholar]
- 19. Metikala S, Venishetty N, Cheppalli N, Jones H, Mounsamy V, Sambandam S. Patients with cirrhosis have higher costs of care, longer length of stays, and more perioperative complications following total knee arthroplasty: a national inpatient sample-based study. Cureus. 2023;15(10):e47317. doi: 10.7759/cureus.47317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Cheppalli NS, Metikala S, Beale J, Mounsamy V, Sambandam S. Total hip arthroplasty in cirrhosis is associated with increased complications during the hospital stay, length of stay, and cost of care: a propensity matched database study. Arch Bone Jt Surg. 2023;11(5):330–336. [PMC free article] [PubMed] [Google Scholar]
- 21. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol. 2010;25(9):603–605. doi: 10.1007/s10654-010-9491-z. [DOI] [PubMed] [Google Scholar]
- 22. Kohn MD, Wolock CJ, Poulson IJ, Fernando ND. A meta-analysis of postoperative outcomes of patients with and without chronic hepatitis C undergoing primary total hip or knee arthroplasty. EFORT Open Rev. 2023;8(4):180–188. doi: 10.1530/EOR-22-0117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Lan R, Stiles ER, Ward SA, Lajam CM, Bosco JA. Patients with moderate to severe liver cirrhosis have significantly higher short-term complication rates following total knee arthroplasty: a retrospective cohort study. J Arthroplasty. 2024;39(7):1736–1740. doi: 10.1016/j.arth.2024.01.039. [DOI] [PubMed] [Google Scholar]
- 24. Belt M, Robben B, Smolders JMH, Schreurs BW, Hannink G, Smulders K. A mapping review on preoperative prognostic factors and outcome measures of revision total knee arthroplasty. Bone Jt Open. 2023;4(5):338–356. doi: 10.1302/2633-1462.45.BJO-2022-0157.R1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Reinhard J, Lang S, Walter N, et al. In-hospital mortality of patients with periprosthetic joint infection. Bone Jt Open. 2024;5(4):367–373. doi: 10.1302/2633-1462.54.BJO-2023-0162.R1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Bhanushali A, Tran L, Nairne-Nagy J, et al. Patient-related predictors of treatment failure after two-stage total hip arthroplasty revision for periprosthetic joint infection: a systematic review and meta-analysis. J Arthroplasty. 2024;39(9):2395–2402. doi: 10.1016/j.arth.2024.04.053. [DOI] [PubMed] [Google Scholar]
- 27. Tiberi JV, Hansen V, El-Abbadi N, Bedair H. Increased complication rates after hip and knee arthroplasty in patients with cirrhosis of the liver. Clin Orthop Relat Res. 2014;472(9):2774–2778. doi: 10.1007/s11999-014-3681-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Moon YW, Kim YS, Kwon SY, Kim SY, Lim SJ, Park YS. Perioperative risk of hip arthroplasty in patients with cirrhotic liver disease. J Korean Med Sci. 2007;22(2):223–226. doi: 10.3346/jkms.2007.22.2.223. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Onochie E, Kayani B, Dawson-Bowling S, Millington S, Achan P, Hanna S. Total hip arthroplasty in patients with chronic liver disease: a systematic review. SICOT J. 2019;5:40. doi: 10.1051/sicotj/2019037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Noor MT, Manoria P. Immune dysfunction in cirrhosis. J Clin Transl Hepatol. 2017;5(1):50–58. doi: 10.14218/JCTH.2016.00056. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Gonzalez Gutierrez F, Sun J, Sambandam S. Risk of blood transfusion following total hip arthroplasty: a national database study from 367,894 patients. J Orthop. 2023;46:112–116. doi: 10.1016/j.jor.2023.10.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Newman KL, Johnson KM, Cornia PB, Wu P, Itani K, Ioannou GN. Perioperative evaluation and management of patients with cirrhosis: risk assessment, surgical outcomes, and future directions. Clin Gastroenterol Hepatol. 2020;18(11):2398–2414. doi: 10.1016/j.cgh.2019.07.051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. López-Larramona G, Lucendo AJ, González-Castillo S, Tenias JM. Hepatic osteodystrophy: an important matter for consideration in chronic liver disease. World J Hepatol. 2011;3(12):300–307. doi: 10.4254/wjh.v3.i12.300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Hodgson SF, Dickson ER, Wahner HW, Johnson KA, Mann KG, Riggs BL. Bone loss and reduced osteoblast function in primary biliary cirrhosis. Ann Intern Med. 1985;103(6 Pt 1):855–860. doi: 10.7326/0003-4819-103-6-855. [DOI] [PubMed] [Google Scholar]
- 35. Barba R, Gonzalvez-Gasch A, Joya Seijo D, et al. Venous thromboembolism in patients with liver diseases. J Thromb Haemost. 2018;16(10):2003–2007. doi: 10.1111/jth.14255. [DOI] [PubMed] [Google Scholar]
- 36. Yang ZJ, Costa KA, Novelli EM, Smith RE. Venous thromboembolism in cirrhosis. Clin Appl Thromb Hemost. 2014;20(2):169–178. doi: 10.1177/1076029612461846. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Enger C, Forssen UM, Bennett D, Theodore D, Shantakumar S, McAfee A. Thromboembolic events among patients with hepatitis C virus infection and cirrhosis: a matched-cohort study. Adv Ther. 2014;31(8):891–903. doi: 10.1007/s12325-014-0138-4. [DOI] [PubMed] [Google Scholar]
- 38. da Cruz Renó L, Tustumi F, Waisberg DR, et al. Venous thromboembolism in in-hospital cirrhotic patients: a systematic review. Front Med (Lausanne) 2022;9:1027882. doi: 10.3389/fmed.2022.1027882. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Williams SN, Wolford ML, Bercovitz A. Hospitalization for total knee replacement among inpatients aged 45 and over: United States, 2000-2010. NCHS Data Brief. 2015;2015(210):1–8. [PubMed] [Google Scholar]
- 40. Pugh RN, Murray-Lyon IM, Dawson JL, Pietroni MC, Williams R. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg. 1973;60(8):646–649. doi: 10.1002/bjs.1800600817. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data generated or analyzed during this study are included in the published article and/or in the supplementary material.