Simple Blood Score Predicts Failure and Death After Joint Replacement Infection Surgery

When a hip or knee replacement becomes infected, the stakes are extraordinarily high. Patients face repeated operations, months of intravenous antibiotics, prolonged hospitalisation, and a real chance of losing the implant or, in the worst cases, their lives. Now a team of Spanish researchers reports that a simple, inexpensive blood test score, calculated before surgery from routine laboratory values, may identify which of these patients are most likely to experience treatment failure or die during follow-up. The finding, published in the Journal of Bone and Joint Infection, could reshape how surgeons and infectious disease specialists assess risk before operating on one of orthopaedic surgery’s most feared complications.

The study, led by Víctor Otero-Naveiro of the Department of Orthopaedic Surgery and Traumatology at University Hospital Lucus Augusti in Lugo, Spain, focused on acute periprosthetic joint infection, or PJI, an infection that develops around an artificial hip or knee joint. PJI affects roughly one percent of patients undergoing contemporary joint replacement, and its burden is likely underestimated because debridement procedures are not consistently captured in national registries. Infection is now a leading cause of failure after total knee arthroplasty, accounting for 21.6 percent of knee revisions and 18.2 percent of hip revisions in recent registry-based analyses. More than 20,000 PJIs arising from primary arthroplasties performed in Europe in 2019 were estimated to account for 346 million euros in reimbursement payments, with septic revision costing up to three times more than aseptic revision.

The tool at the heart of the new research is the CONUT score, short for Controlling Nutritional Status, originally developed in 2005 as a hospital-based nutritional screening instrument. The score combines three values that are already measured in almost every preoperative blood panel: serum albumin, which reflects protein reserve; total lymphocyte count, which captures immune-nutritional status; and total cholesterol, which signals caloric depletion. Each component contributes points, producing a total score from 0 to 12 that is categorised as normal, mild, moderate, or severe. In this study, the researchers dichotomised patients as low CONUT, below 5, or high CONUT, at 5 or above, corresponding to the moderate-to-severe threshold used in previous PJI outcome literature.

The retrospective cohort drew on the institutional database of the Bone and Joint Infection Unit at the tertiary referral centre, screening all consecutive adult patients surgically treated for acute hip or knee PJI between 1 January 2013 and 30 November 2024. Infection was defined according to the European Bone and Joint Infection Society criteria, with acute postoperative infection defined as occurring within three months of the index arthroplasty with symptoms lasting no more than three weeks, and acute haematogenous infection defined as symptoms of up to three weeks in a previously well-functioning prosthesis. Of 169 patients screened, 59 were excluded because at least one CONUT component was unavailable, leaving 90 patients for analysis: 16 hip hemiarthroplasties, 39 total hip arthroplasties, and 35 total knee arthroplasties, with a median follow-up of 46.5 months.

The results were striking. Forty-two of the 90 patients, or 46.7 percent, had a high CONUT score before surgery. These patients were older, more frequently aged 80 or above, more often classified as McPherson host grade C, and had higher preoperative C-reactive protein. Crucially, they were also more likely to have Staphylococcus aureus infection and positive blood cultures, while coagulase-negative staphylococci were more common in the low-CONUT group. In the overall cohort, S. aureus infection was strongly associated with positive blood cultures, occurring in 45.5 percent of such cases versus 14.0 percent, a pattern consistent with a more systemic, invasive infection phenotype.

Outcomes were assessed using the Musculoskeletal Infection Society Outcome Reporting Tool, a structured classification that distinguishes failure, defined as septic revision or reoperation, early aseptic revision, amputation, resection arthroplasty, arthrodesis, or a retained spacer, from total failure, which additionally includes death within one year of starting PJI treatment, and from overall mortality from any cause. High CONUT was associated with failure in 35.7 percent of patients versus 10.4 percent in the low group, with total failure in 57.1 percent versus 12.5 percent, and with overall mortality in 61.9 percent versus 22.9 percent. All three differences were statistically significant, with p values of 0.004 or lower.

The associations survived rigorous statistical adjustment. In multivariable logistic regression, high CONUT remained independently associated with failure, with an odds ratio of 4.24, and with total failure, with an odds ratio of 9.57, meaning high-CONUT patients had nearly ten times the adjusted odds of either failing treatment or dying within a year. In the Cox proportional hazards model for overall mortality, high CONUT carried a hazard ratio of 2.73, alongside older age, diabetes mellitus, and a fracture or revision indication. Kaplan-Meier analyses confirmed significantly lower failure-free survival, total-failure-free survival, and overall survival in the high-CONUT group, with median overall survival of 42.3 months in the high-CONUT group and not yet reached in the low-CONUT group. An exploratory sensitivity analysis treating CONUT as a continuous variable found that each one-point increase in the score raised the hazard of death by 34 percent, reinforcing a dose-response relationship.

Perhaps the most scientifically interesting aspect of the study is the authors’ caution about what the CONUT score actually measures in this setting. They argue that in acute PJI it should not be interpreted as a formal diagnosis of malnutrition. According to the Global Leadership Initiative on Malnutrition framework, malnutrition requires both phenotypic and aetiological criteria rather than serum abnormalities alone, and updated European Society for Clinical Nutrition and Metabolism guidance explicitly states that albumin is prognostic but not a nutritional marker per se. Because CONUT components were measured after symptom onset and close to surgery, systemic inflammation, bacteraemia, metabolic stress, and catabolic activation may all depress albumin, lymphocyte, and cholesterol levels. Hypoalbuminaemia may reflect the hepatic acute-phase response, lymphopenia may accompany immune dysregulation and bacteraemia, and low cholesterol has been linked to sepsis severity and mortality. In acute infection, the score may therefore function best as a composite host-response marker integrating nutritional reserve, inflammation, immune dysfunction, and catabolic stress, rather than a pure nutrition gauge.

Notably, conventional inflammatory biomarkers performed less well. C-reactive protein, the CRP-to-albumin ratio, and the neutrophil-to-lymphocyte ratio showed no consistent association with outcomes across the study, echoing earlier findings that isolated inflammatory indices may provide a weaker prognostic signal than composite host-status measures. Diabetes mellitus emerged as the other consistent independent predictor, associated with all three outcomes, a biologically plausible result given that hyperglycaemia impairs neutrophil chemotaxis, phagocytosis, and bactericidal activity. The authors suggest that CONUT should complement rather than replace established PJI-specific host and complexity classifications such as the Joint-Specific BACH classification, adding an objective laboratory-based dimension of host vulnerability.

The study has limitations that warrant careful interpretation. The retrospective, single-centre design, the modest sample size of 90 patients, the limited number of events, and residual confounding all constrain the strength of the conclusions, and surgical treatment was heterogeneous and non-randomised, so treatment-specific comparisons are descriptive rather than causal. The exclusion of 59 patients with missing CONUT components may have introduced selection bias, although included and excluded patients showed no significant baseline differences. Even so, the implications are compelling: three values drawn from blood tests that almost every surgical patient already receives may help clinicians identify vulnerable patients before they enter the operating theatre, supporting better-informed decisions about surgical strategy, intensity of monitoring, and communication of risk. The authors call for larger multi-centre studies to validate the findings and to determine whether targeted preoperative optimisation strategies can actually modify the outcomes that a high CONUT score appears to foreshadow. If confirmed, a score that costs almost nothing to calculate could become a routine part of preoperative risk stratification for one of the costliest and most dangerous complications in modern joint surgery.

Subject of Research: Preoperative CONUT score as a predictor of treatment failure and mortality in acute periprosthetic joint infection

Article Title: High preoperative CONUT score is associated with treatment failure and mortality after surgery for acute periprosthetic joint infection: a single-centre retrospective cohort study

Article References: Otero-Naveiro, V., Ayuso-García, B., Pena-Paz, S., Rión-Giraldo, R. A., García-Pais, M. J., & López-Pardo Pardo, A. (2026). High preoperative CONUT score is associated with treatment failure and mortality after surgery for acute periprosthetic joint infection: a single-centre retrospective cohort study. Journal of Bone and Joint Infection, 11(5), 613-622. https://doi.org/10.5194/jbji-11-613-2026

Image Credits: AI Generated

DOI: 10.5194/jbji-11-613-2026

Keywords: periprosthetic joint infection, CONUT score, hip arthroplasty, knee arthroplasty, treatment failure, mortality, malnutrition screening, serum albumin, lymphocyte count, blood cultures, Staphylococcus aureus, risk stratification