Antibiotic Therapy Strategies for Treating Gram-Negative Severe Infections in the Critically Ill: A Narrative Review.

INTRODUCTION: Not enough data exist to inform the optimal duration and type of antimicrobial therapy against GN infections in critically ill patients. METHODS: Narrative review based on a literature search through PubMed and Cochrane using the following keywords: "multi-drug resistant (MDR)", "extensively drug resistant (XDR)", "pan-drug-resistant (PDR)", "difficult-to-treat (DTR) Gram-negative infection," "antibiotic duration therapy", "antibiotic combination therapy" "antibiotic monotherapy" "Gram-negative bacteremia", "Gram-negative pneumonia", and "Gram-negative intra-abdominal infection". RESULTS: Current literature data suggest adopting longer (≥10-14 days) courses of synergistic combination therapy due to the high global prevalence of ESBL-producing (45-50%), MDR (35%), XDR (15-20%), PDR (5.9-6.2%), and carbapenemases (CP)/metallo-ß-lactamases (MBL)-producing (12.5-20%) Gram-negative (GN) microorganisms (i.e., Klebsiella pneumoniae, Pseudomonas aeruginosa, and Acinetobacter baumanii). On the other hand, shorter courses (≤5-7 days) of monotherapy should be limited to treating infections caused by GN with higher (≥3 antibiotic classes) antibiotic susceptibility. A general approach should be based on (i) third or further generation cephalosporins ± quinolones/aminoglycosides in the case of MDR-GN; (ii) carbapenems ± fosfomycin/aminoglycosides for extended-spectrum ß-lactamases (ESBLs); and (iii) the association of old drugs with new expanded-spectrum ß-lactamase inhibitors for XDR, PDR, and CP microorganisms. Therapeutic drug monitoring (TDM) in combination with minimum inhibitory concentration (MIC), bactericidal vs. bacteriostatic antibiotics, and the presence of resistance risk predictors (linked to patient, antibiotic, and microorganism) should represent variables affecting the antimicrobial strategies for treating GN infections. CONCLUSIONS: Despite the strategies of therapy described in the results, clinicians must remember that all treatment decisions are dynamic, requiring frequent reassessments depending on both the clinical and microbiological responses of the patient.

Laboratory predictors of death from coronavirus disease 2019 (COVID-19) in the area of Valcamonica, Italy.

Background Comprehensive information has been published on laboratory tests which may predict worse outcome in Asian populations with coronavirus disease 2019 (COVID-19). The aim of this study is to describe laboratory findings in a group of Italian COVID-19 patients in the area of Valcamonica, and correlate abnormalities with disease severity. Methods The final study population consisted of 144 patients diagnosed with COVID-19 (70 who died during hospital stay and 74 who survived and could be discharged) between March 1 and 30, 2020, in Valcamonica Hospital. Demographical, clinical and laboratory data were collected upon hospital admission and were then correlated with outcome (i.e. in-hospital death vs. discharge). Results Compared to patients who could be finally discharged, those who died during hospital stay displayed significantly higher values of serum glucose, aspartate aminotransferase (AST), creatine kinase (CK), lactate dehydrogenase (LDH), urea, creatinine, high-sensitivity cardiac troponin I (hscTnI), prothrombin time/international normalized ratio (PT/INR), activated partial thromboplastin time (APTT), D-dimer, C reactive protein (CRP), ferritin and leukocytes (especially neutrophils), whilst values of albumin, hemoglobin and lymphocytes were significantly decreased. In multiple regression analysis, LDH, CRP, neutrophils, lymphocytes, albumin, APTT and age remained significant predictors of in-hospital death. A regression model incorporating these variables explained 80% of overall variance of in-hospital death. Conclusions The most important laboratory abnormalities described here in a subset of European COVID-19 patients residing in Valcamonica are highly predictive of in-hospital death and may be useful for guiding risk assessment and clinical decision-making.

Control and prevention measures for legionellosis in hospitals: A cross-sectional survey in Italy.

Risk assessment, environmental monitoring, and the disinfection of water systems are the key elements in preventing legionellosis risk. The Italian Study Group of Hospital Hygiene of the Italian Society of Hygiene, Preventive Medicine, and Public Health and the Italian Multidisciplinary Society for the Prevention of Health Care-Associated Infections carried out a national cross-sectional survey to investigate the measures taken to prevent and control legionellosis in Italian hospitals. A multiple-choice questionnaire was developed, comprising 71 questions regarding hospital location, general characteristics, clinical and environmental surveillance, and control and preventive measures for legionellosis in 2015. Overall, 739 hospitals were enrolled from February to June 2017, and 178 anonymous questionnaires were correctly completed and evaluated (response rate: 24.1%). The survey was conducted using the SurveyMonkey® platform, and the data were analyzed using Stata 12 software. Of the participating hospitals, 63.2% reported at least one case of legionellosis, of which 28.2% were of proven nosocomial origin. The highest case numbers were reported in the Northern Italy, in hospitals with a pavilion structure or cooling towers, and in hospitals with higher numbers of beds, wards and operating theaters. Laboratory diagnosis was performed using urinary antigen testing alone (31.9%), both urinary antigen testing and single antibody titer (17.8%), or with seroconversion also added (21.5%). Culture-based or molecular investigations were performed in 28.8% and 22.1% of the clinical specimens, respectively. The water systems were routinely tested for Legionella in 97.4% of the hospitals, 62% of which detected a positive result (> 1000 cfu/L). Legionella pneumophila serogroup 2-15 was the most frequently isolated species (58.4%). The most common control measures were the disinfection of the water system (73.7%), mostly through thermal shock (37.4%) and chlorine dioxide (34.4%), and the replacement (69.7%) or cleaning (70.4%) of faucets and showerheads. A dedicated multidisciplinary team was present in 52.8% of the hospitals, and 73% of the hospitals performed risk assessment. Targeted training courses were organized in 36.5% of the hospitals, involving nurses (30.7%), physicians (28.8%), biologists (21.5%), technicians (26.4%), and cleaners (11%). Control and prevention measures for legionellosis are present in Italian hospitals, but some critical aspects should be improved. More appropriate risk assessment is necessary, especially in large facilities with a high number of hospitalizations. Moreover, more sensitive diagnostic tests should be used, and dedicated training courses should be implemented.

Variable performance of models for predicting methicillin-resistant Staphylococcus aureus carriage in European surgical wards.

BACKGROUND: Predictive models to identify unknown methicillin-resistant Staphylococcus aureus (MRSA) carriage on admission may optimise targeted MRSA screening and efficient use of resources. However, common approaches to model selection can result in overconfident estimates and poor predictive performance. We aimed to compare the performance of various models to predict previously unknown MRSA carriage on admission to surgical wards. METHODS: The study analysed data collected during a prospective cohort study which enrolled consecutive adult patients admitted to 13 surgical wards in 4 European hospitals. The participating hospitals were located in Athens (Greece), Barcelona (Spain), Cremona (Italy) and Paris (France). Universal admission MRSA screening was performed in the surgical wards. Data regarding demographic characteristics and potential risk factors for MRSA carriage were prospectively collected during the study period. Four logistic regression models were used to predict probabilities of unknown MRSA carriage using risk factor data: "Stepwise" (variables selected by backward elimination); "Best BMA" (model with highest posterior probability using Bayesian model averaging which accounts for uncertainty in model choice); "BMA" (average of all models selected with BMA); and "Simple" (model including variables selected >50% of the time by both Stepwise and BMA approaches applied to repeated random sub-samples of 50% of the data). To assess model performance, cross-validation against data not used for model fitting was conducted and net reclassification improvement (NRI) was calculated. RESULTS: Of 2,901 patients enrolled, 111 (3.8%) were newly identified MRSA carriers. Recent hospitalisation and presence of a wound/ulcer were significantly associated with MRSA carriage in all models. While all models demonstrated limited predictive ability (mean c-statistics <0.7) the Simple model consistently detected more MRSA-positive individuals despite screening fewer patients than the Stepwise model. Moreover, the Simple model improved reclassification of patients into appropriate risk strata compared with the Stepwise model (NRI 6.6%, P = .07). CONCLUSIONS: Though commonly used, models developed using stepwise variable selection can have relatively poor predictive value. When developing MRSA risk indices, simpler models, which account for uncertainty in model selection, may better stratify patients' risk of unknown MRSA carriage.

Pneumonia in a cardiothoracic intensive care unit: incidence and risk factors.

OBJECTIVE: The purpose of this study was to determine the incidence, risk factors, and pathogens causing pneumonia in a cardiothoracic intensive care unit (CTICU). DESIGN: A prospective study. SETTING: "Civili Hospital," Brescia, Italy. PARTICIPANTS: One hundred forty consecutive patients in the CTICU for more than 24 hours from October 1, 2006, to September 30, 2007. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Demographic variables and intrinsic and extrinsic risk factors were analyzed with univariate and multivariate analysis. One hundred forty patients were studied, 128 (91.4%) were surgical and 12 (8.5%) were medical. Cumulative incidence of pneumonia was 28.6% (n = 40); 62.5% (n = 25) had ventilator-associated pneumonia (VAP) and 37.5% (n = 15) had non-VAP. The most common isolated pathogens were Pseudomonas aeruginosa (n = 15), Staphylococcus aureus (n = 5), Escherichia coli (n = 4), and Klebsiella pneumoniae (n = 3). Mortality was 22.2% (n = 31), with 54.8% (n = 17) of patients with pneumonia leading to mortality during CTICU stay (p = 0.0006). On multivariate analysis, independent risk factors for pneumonia were each point of the Sequential Organ Failure Assessment score at CTICU admission (p = 0.006, odds ratio [OR] = 1.39, confidence interval [CI] = 1.09-1.76), every day of mechanical ventilation (p = 0.049, OR = 1.08, CI = 1.00-1.18), noninvasive mechanical ventilation (NIMV) (p = 0.014, OR = 4.83, CI = 1.37-17.03), and bronchoscopy (p = 0.002, OR = 8.14, CI = 2.10-31.55). CONCLUSIONS: Pneumonia is a common complication in the CTICU, and the authors recommend the following: the removal of the endotracheal tube as soon as possible, the minimal use of a bronchoscope and only in cases of bronchial obstruction, and the use of NIMV.

Adherence to hand hygiene in an Italian long-term care facility.

In an Italian long-term-care facility (LTCF), we observed a 17.5% adherence to hand hygiene (HH), as well as 47.5% rate of glove use. Performing a procedure at high risk for cross-transmission of germs was the factor most strongly associated with noncompliance (odds ratio = 13.3; 95% confidence interval = 6.2 to 28.8; P < .0001). No significant differences in compliance related to health care worker category were found. Adherence to HH in the LTCF was similar to that found in a rehabilitation medicine unit of an acute care hospital (15.8%) but significantly lower than that reported in an infectious disease unit (53.7%; P < .0001). Our findings indicate that compliance with HH is a similar problem in LTCFs as in acute care facilities.

Hand hygiene and glove use behavior in an Italian hospital.

In an Italian hospital, we observed that hand hygiene was performed in 638 (19.6%) of 3,253 opportunities, whereas gloves were worn in 538 (44.2%) 1,218 of opportunities. We observed an inverse correlation between the intensity of care and the rate of hand hygiene compliance (R2=0.057; P<.001), but no such association was observed for the rate of glove use compliance (R2=0.014; P=.078). Rates of compliance with hand hygiene and glove use recommendations follow different behavioral patterns.

Adherence and plasma drug concentrations are predictors of confirmed virologic response after 24-week salvage highly active antiretroviral therapy.

Data from 197 patients for whom highly active antiretroviral therapy (HAART) failed, who started a new regimen chosen under the guide of resistance testing results interpreted by experts, were retrospectively studied, provided that at least 2 determinations of adherence and plasma drug concentrations were performed during the follow-up. Univariate and multivariable logistic regression analyses were conducted, using confirmed virologic response at week 24 as outcome measure (i.e., achievement of undetectable HIV plasma viral load at any time point before week 24 and its maintenance up to week 24). Suboptimal drug concentrations (odds ratio [OR]: 0.3; 95% confidence interval [CI] 0.2-0.7; p = 0.006) and suboptimal adherence (OR: 0.4; 95% CI 0.2-0.8; p = 0.014) were both negative independent predictors of sustained virologic response, while the use of boosted protease inhibitor-containing regimens resulted to be protective (OR: 2.4; 95% CI 1.1-5.3; p = 0.032).

Influence of viral chronic hepatitis co-infection on plasma drug concentrations and liver transaminase elevations upon therapy switch in HIV-positive patients.

It is still controversial whether viral hepatitis co-infection can influence antiretroviral plasma drug concentrations and whether drug concentrations are correlated with liver enzyme elevations during highly active antiretroviral therapy. An analysis of data from a cohort of 220 human immunodeficiency virus (HIV)-infected patients was conducted. Univariate and multivariate logistic analyses were performed to identify predictors of plasma drug concentrations. The association of transaminase elevation with higher plasma drug concentrations was explored following stratification of patients into HIV monoinfected and hepatitis C virus (HCV) and/or hepatitis B virus (HBV) co-infected groups. Hepatitis co-infections were independently correlated with drug concentrations above the therapeutic cut-offs at Week 1 (P=0.06), Week 4 (P=0.04) and Week 12 (P=0.005). The apparent effect was independent of the possible impact exerted by other variables such as demographics and medication adherence. The incidence of relevant hypertransaminasaemia was low. Patients with hepatitis co-infections had higher rates of transaminase elevation than monoinfected HIV patients; however, risk of transaminase elevation was not associated with drug concentrations. The presence of HCV and/or HBV co-infections correlated with higher plasma drug concentrations, although it did not appear to influence hepatotoxicity risk.

Vaginal colonization with Candida spp. in human immunodeficiency virus-infected women: a cohort study.

We have conducted a longitudinal study on factors associated with candidal vaginal colonization, a precursor of vaginitis, in a cohort of HIV-infected women in Italy. All consecutive women attending a single, tertiary care clinical site were offered free screening for sexually transmitted infections and genital disorders every 6-12 months. Candidal vaginal colonization was defined as a positive culture for Candida spp. in an asymptomatic woman. From January 1998 to July 2002 we analysed 214 women. The baseline prevalence of candidal vaginal colonization was 16.8%. In the logistic regression analysis, the time since HIV infection > or =36 months (odds ratio [OR] = 0.18, 95% confidence interval [CI] 0.016-0.53, P = 0.002) and a plasma viral load > or =10,000 copies/mL (OR = 3.9, 95% CI 1.03-14.9, P = 0.045) were independently associated with candidal colonization. Among 130 women who were followed for a mean period of 24 months, the incidence of vaginal colonization was 10.7/100 women-years. In the Cox regression analysis, a CD4(+) T-lymphocytes count <100 cells/microL during the follow-up was associated with an increased risk of candidal vaginal colonization (OR = 4.45, C.I. = 1.20-16.81, P = 0.03). Risk of candidal vaginal colonization episodes in HIV-infected women significantly increase when CD4(+) T-lymphocytes are less than 100.

Prediction of early and confirmed virological response by genotypic inhibitory quotients for lopinavir in patients naïve for lopinavir with limited exposure to previous protease inhibitors.

OBJECTIVE: To determine the impact of genotypic inhibitory quotient (GIQ) for lopinavir (LPV) in patients failing HAART with limited antiretroviral exposure. DESIGN: Retrospective analysis of a prospective trial. METHODS: Lopinavir GIQ was calculated as the ratio between the mean trough concentration (C(trough)) and the number of protease mutations using eight different HIV drug resistance mutation lists or algorithms. Early (by week 12) and confirmed (up to week 24) virological response (HIV-RNA< 400 copies/mL, ECVR) was used as dependent variable in logistic regression model. RESULTS: Seventy-one of 109 (65%) patients achieved ECVR. At multivariable logistic regression analysis, each mug/mL increase of GIQ was correlated with increasing probability of ECVR as far as the following mutations were computed: multi-protease inhibitor (PI) associated mutations listed by IAS (OR=1.17; 95% CI=0.99-1.39; P=0.058), mutations associated with LPV resistance by ANRS algorithm (OR=1.21; 95% CI=1.02-1.44; P=0.03), major mutations associated with LPV resistance by Stanford database (OR=1.16; 95% CI=1-1.35; P=0.05), and the whole set of mutations associated with LPV resistance in the same database (OR=1.22; 95% CI=1.02-1.46; P=0.03). Using ROC curve method, a specific threshold GIQ was assessed, above which this parameter could predict ECVR with the highest sensitivity (74.6% with GIQ obtained through Stanford LPV mutations) or specificity (89.5% with GIQ obtained through ANRS LPV mutations). CONCLUSIONS: Our results suggest that increasing GIQ can improve virological outcome even in patients with limited exposure to PIs. Further studies are necessary to understand what HIV protease mutations should be considered and whether such mutations should be weighted differently to improve LPV GIQ predictive value.

Early virological failure after tenofovir + didanosine + efavirenz combination in HIV-positive patients upon starting antiretroviral therapy.

A prospective, randomized pilot trial was conducted in naive patients comparing three different combinations: zidovudine+lamivudine+lopinavir/ritonavir (arm A) versus tenofovir+lamivudine+efavirenz (arm B) versus tenofovir+didanosine+efavirenz (arm C). HIV-RNA slope (days 1, 3, 7, 14 and 28) was slower in arm C with respect to arm B (P < 0.0001). Seven out of eight patients (87.5%) reached undetectable HIV-RNA by week 28 in arm A, 10/10 (100%) in arm B and 6/10 (60%) in arm C. Among arm C patients who failed at week 4, one HIV isolate showed 67N and 219Q, and another one showed 210F and 215D substitutions in the HIV reverse transcriptase gene at baseline, respectively. Non-nucleoside reverse transcriptase inhibitor resistance-related mutations appeared first, followed by 65R mutations in all cases. Efavirenz AUC(0-24) values were lower in arm C with respect to arm B, especially in patients who failed early. A high virological failure rate after tenofovir+didanosine+efavirenz correlated with a slower HIV-RNA decrease and a peculiar accumulation of resistance mutations. A constellation of factors could be correlated with early failure events in patients receiving this combination such as resistance mutations or polymorphisms present at baseline, low CD4+ T-cell count or advanced disease and unexpectedly low efavirenz plasma levels.

Exploratory analysis for the evaluation of lopinavir/ritonavir-versus efavirenz-based HAART regimens in antiretroviral-naive HIV-positive patients: results from the Italian MASTER Cohort.

OBJECTIVE: This retrospective longitudinal cohort study compared the virological and immunological responses to highly active antiretroviral therapy containing either efavirenz or lopinavir/ritonavir in previously antiretroviral-naive HIV-infected patients. PATIENTS AND METHODS: A total of 472 patients were selected (348 efavirenz and 124 lopinavir/ritonavir). The primary endpoint of this study was virological success (HIV RNA <50 copies/mL). The immunological response was assessed on the basis of either CD4+ T cell count variations (absolute and percentage) with respect to baseline values or categorical endpoints (defined as either a CD4+ T cell increase of > or =1;50 cells/mm(3) at week 24 or of > or =1;75 cells/mm(3) at week 48). RESULTS: At intention-to-treat (ITT) analysis, the adjusted odds ratio of virological success for patients who started lopinavir/ritonavir, compared with those who started efavirenz, was 0.54 (95% CI: 0.33-0.89, P = 0.016) at week 24 and 0.40 (95% CI: 0.33-0.89, P = 0.002) at week 48. However, patients receiving lopinavir/ritonavir had a more pronounced CD4+ T cell recovery, demonstrating both a mean absolute and percentage increase up to week 48 (MANOVA P < 0.0001). CONCLUSIONS: Although comparisons of drug efficacy in non-randomized studies should be viewed with caution, from a virological point of view efavirenz-containing regimens performed as well (on-treatment analysis) or better (ITT analysis) than those containing lopinavir/ritonavir. In contrast, immunological outcome appeared to favour lopinavir/ritonavir.

A randomized controlled trial to evaluate antiretroviral salvage therapy guided by rules-based or phenotype-driven HIV-1 genotypic drug-resistance interpretation with or without concentration-controlled intervention: the Resistance and Dosage Adapted Regimens (RADAR) study.

BACKGROUND: It is not well defined whether concentration-controlled intervention (CCI) and rules-based human immunodeficiency virus (HIV) type 1 genotype drug-resistance interpretation (GI) or virtual phenotype drug-resistance interpretation (VPI) may improve the outcome of HIV salvage therapy. METHODS: In a prospective, randomized, controlled trial, patients were randomized (on a factorial basis) to change treatment after either GI or VPI, and they then were further randomized to the control arm (no CCI) or the CCI arm. Protease inhibitor (PI) and nonnucleoside reverse-transcriptase inhibitor (NNRTI) trough concentration (Ctrough) values were determined at weeks 1, 4, 12, and 24 of the study. RESULTS: Among 230 patients, virological benefit (defined by an HIV RNA load of <400 copies/mL at week 24) was not statistically different, either between patients in the GI and VPI arms or between patients in the CCI and control arms. A virological benefit was found for patients in the CCI arm, compared with patients in the control arm, but this benefit was not statistically significant (56.8% vs. 64.3% at week 4 and 63.6% vs. 74% at week 12). Dosage adaptation was possible for only a fraction of patients, because of low rates of treatment adherence or patient refusal to increase dosages. In the logistic regression analysis, independent predictors of virological response at week 24 were a PI Ctrough value and/or an NNRTI Ctrough value in the higher quartiles (or above cutoff levels) and a low number of PIs previously received. Moreover, receipt of a regimen that contained PIs boosted with ritonavir was an independent predictor of virological response. CONCLUSIONS: The present study did not support the routine use of CCI for patients undergoing salvage treatment, probably as a result of existing difficulties associated with its clinical application. However, a higher Ctrough value appeared to be correlated with treatment response. No major differences were found between VPI or GI when they are used together with expert advice for the selection of salvage treatment combinations.

Immune correlates of virological response in HIV-positive patients after highly active antiretroviral therapy (HAART).

Correlates of immune reconstitution after highly active antiretroviral therapy (HAART) are not completely understood, in particular as far as viro-immunological discordant responses are concerned. HIV-positive patients on stable HAART for > or = 1 year were recruited. Viro-immunological responses were categorized according to positive or negative area under the curve (AUC) variations for HIV plasma viral load (pVL) and CD4+ T-cell counts measured at least every 4 months. The following parameters were evaluated: lymphocyte spontaneous apoptosis (LSA), intracellular Bcl-2 expression in both CD4-CD45RA+ and CD4-CD45R0+, IL-7 and IL-15 plasma concentrations, and lymphocyte TRECs levels. Sixty-one patients were enrolled. A significant inverse correlation was found between CD4+ T-cell count and pVL AUC (r = 0.45; p = 0.0003). Patients with pVL response had higher levels of Bcl-2 in CD4-CD45R0+ (mean 65,409 MESF vs. 54,018 MESF; p = 0.089) and higher IL-15 (mean 1.34 pg/mL vs. 1.05 pg/mL; p = 0.069, respectively). Higher LSA and lower TRECs levels were found in viro-immunological non-responder patients with respect to those who had viro-immunological response (mean 24.84% vs. 14.89%; p = 0.01, and mean 17,796 copies/10(6) cells vs. 29,251 copies/10(6) cells; p = 0.68, respectively). Virological suppression may allow Bcl-2 and IL-15 hyperexpression during incomplete immune-reconstitution phase, while more complete immune reconstitution appeared to be marked by both high TRECs and low LSA levels, possibly indicating both central and peripheral CD4+ T-cell repopulations at this stage.

Drug resistance mutations and newly recognized treatment-related substitutions in the HIV-1 protease gene: prevalence and associations with drug exposure and real or virtual phenotypic resistance to protease inhibitors in two clinical cohorts of antiretroviral experienced patients.

This study aimed at identifying HIV-1 protease amino acid changes associated with protease inhibitor (PI) exposure and susceptibility. New amino acid substitutions were correlated with the number of experienced PIs, reaching statistical significance only for those at positions 3, 44, and 74. The correspondence multivariate model demonstrated that > or =3 experienced PIs and substitutions or mutations at positions 3, 46, 54, 73, 74, and 84 were correlated with PI cross-resistance, including resistance for lopinavir and amprenavir in this cohort of patients who were naive for these drugs.

HIV susceptibility to amprenavir: phenotype-based versus rules-based interpretations.

OBJECTIVES: The objective was to study genotypic correlates of discordant interpretations of amprenavir (APV) resistance between a rules-based algorithm and either recombinant phenotype or virtual phenotype. METHODS: HIV resistance mutations found in patients from the GenPheRex study were interpreted with VGI-TRUGENE (version 5.0; VGI) and compared with either recombinant-phenotype (Antivirogram, r-PHT) or virtual-phenotype (Virtual-Phenotype, v-PHT) interpreted through Virco biological cut-offs. RESULTS: Among 180 samples available, 56 (31.1%) were discordant with the observed genotype interpretation results, as a result of being judged as sensitive by r-PHT or v-PHT but resistant by VGI (S/R). Only the I84V mutation was almost invariably found in concordant resistant isolates compared with S/R isolates (60% versus 0%, respectively; P < 0.0001). Notwithstanding this, the number of multi-protease inhibitor-associated mutations (PAMs) was significantly higher in the concordant resistant isolates; the prevalence of >3 PAMs was 56.52% versus 33.93% in R/R and S/R isolates, respectively (P = 0.01). Correspondence analysis confirmed the relevance of PAMs, although additional mutations appeared to be correlated with APV resistance. CONCLUSIONS: The rate of discordance between rules-based and either r-PHT or v-PHT interpretations for APV was high. Mutation I84V and accumulation of >3 PAMs were found to be associated with resistance as interpreted with all systems tested. However, our results indicate that a number of mutations may have an impact on APV resistance, but that they are missed by current interpretation algorithms and this merits further investigations.