Identifying treatment responders using counterfactual modeling and potential outcomes.

Individualizing treatment according to patients' characteristics is central for personalized or precision medicine. There has been considerable recent research in developing statistical methods to determine optimal personalized treatment strategies by modeling the outcome of patients according to relevant covariates under each of the alternative treatments, and then relying on so-called predicted individual treatment effects. In this paper, we use potential outcomes and principal stratification frameworks and develop a multinomial model for left and right-censored data to estimate the probability that a patient is a responder given a set of baseline covariates. The model can apply to RCT or observational study data. This method is based on the monotonicity assumption, which implies that no patients would respond to the control treatment but not to the experimental one. We conduct a simulation study to evaluate the properties of the proposed estimation method. Results showed that the predictions of the probability of being a responder were well calibrated even if we observed variability and a small bias when many parameters were estimated. We finally applied the method to a cohort study on the selection of patients for additional radiotherapy after resection of a soft-tissue sarcoma.

Reporting of heterogeneity of treatment effect in cohort studies: a review of the literature.

BACKGROUND: This article corresponds to a literature review and analyze how heterogeneity of treatment (HTE) is reported and addressed in cohort studies and to evaluate the use of the different measures to HTE analysis. METHODS: prospective cohort studies, in English language, measuring the effect of a treatment (pharmacological, interventional, or other) published among 119 core clinical journals (defined by the National Library of Medicine) in the last 16 years were selected in the following data source: Medline. One reviewer randomly sampled journal articles with 1: 1 stratification by journal type: high impact journals (the New England Journal of Medicine, JAMA, LANCET, Annals of Internal Medicine, BMJ and Plos Medicine) and low impact journal (the remaining journals) to identify 150 eligible studies. Two reviewers independently and in duplicate used standardized piloted forms to screen study reports for eligibility and to extract data. They also used explicit criteria to determine whether a cohort study reported HTE analysis. Logistic regression was used to examine the association of prespecified study characteristics with reporting versus not reporting of heterogeneity of treatment effect. RESULTS: One hundred fifty cohort studies were included of which 88 (58%) reported HTE analysis. High impact journals (Odds Ratio: 3.5, 95% CI: 1.78-7.5; P < 0.001), pharmacological studies (Odds Ratio: 0.26, 95% CI: 0.13-0.51; P < 0.001) and studies published after 2014 (Odds Ratio: 0.5, 95% CI: 0.25-0.97; P = 0.004) were associated with more frequent reporting of HTE. 27 (31%) studies which reported HTE used an interaction test. CONCLUSION: More than half cohort studies report some measure of heterogeneity of treatment effect. Prospective cohort studies published in high impact journals, with large sample size, or studying a pharmacological treatment are associated with more frequent HTE reporting. The source of funding was not associated with HTE reporting. There is a need for guidelines on how to perform HTE analyses in cohort studies.

Patients with Revision Modern Megaprostheses of the Distal Femur Have Improved Disease-Specific and Health-Related Outcomes Compared to Those with Primary Replacements.

We asked whether there would be any difference between primary and revision modern cemented fixed hinge megaprosthesis of the distal femur in function and activity-related outcomes following treatment of a bone tumor. An identical custom-made fixed hinge cemented megaprosthesis with a hydroxyapatite collar was used in all cases. The main outcomes were joint-specific function, disease-specific activity, and health-related quality of life. Implant survival was also evaluated. Patients in the revision group performed slightly better than patients in the primary group on disease-specific (Toronto Extremity Salvage Score, p = 0.033; Musculoskeletal Tumor Society, p = 0.072) and health-related outcomes (Short Form 36 [SF-36] physical component, p = 0.085; SF-36 mental component, p = 0.069) but not on joint-specific outcomes (Knee Society Score, p = 0.94). The cumulative probabilities of revision for any reason were 14.5% (7-25%) at 5 years with no statistically significant difference between primary and revision procedures (p = 0.77). In conclusion, patients undergoing a revision have similar joint-specific functional outcome but improved disease-specific and health-related outcomes. Implant survival are similar between groups.

Optimal post-operative radiation after soft-tissue sarcoma resection is achieved in less than two thirds of cases.

PURPOSE: Local control of soft tissue sarcomas frequently involves adjuvant radiation to the surgical resection. When opting for post-operative radiation, care should be taken that radiation is started within some reasonable time after the surgery. We were interested to the proportion of patients who did not recieve optimal post-operative radiation and the variables associated. METHODS: We retrospectively analyzed a series of 77 patients operated on for an extremity soft-tissue sarcoma and due for post-operative radiotherapy. Patients were considered to have received radiation optimally if radiation was started within 12 weeks of surgery. Variables associated with not receiving radiation optimally were looked for using univariable and multivariable regression models. RESULTS: Overall, 26 patients (34%; 95% CI: 23-45%) did not receive radiation optimally. Twenty (26%) did not start radiation within the 12-weeks mark and six (8%) could not have radiation at all. The main reason identified for not receiving radiation on time was the occurrence of a wound complication (14 (54%) patients). An increased body mass index (OR: 1.14; 95% CI: 1.02-1.26; p = 0.02) and an older age (OR: 1.55; 95% CI: 1.18-2.14; p = 0.04; of note, the OR are for a 10 year change) were significantly associated with not receiving radiation optimally. Patients with a social fragility (p = 0.04), metastatic spread at presentation (p = 0.04), and increased duration of surgery (p = 0.02) were more likely to develop a wound complication. CONCLUSIONS: About 34% of patients do not receive optimal post-operative radiation treatment. Older and obese patients have a higher risk of not receiving radiation optimally. The decision for pre- or post-operative radiation should account for these findings.

Using Bayesian statistics to estimate the likelihood a new trial will demonstrate the efficacy of a new treatment.

BACKGROUND: The common frequentist approach is limited in providing investigators with appropriate measures for conducting a new trial. To answer such important questions and one has to look at Bayesian statistics. METHODS: As a worked example, we conducted a Bayesian cumulative meta-analysis to summarize the benefit of patient-specific instrumentation on the alignment of total knee replacement from previously published evidence. Data were sourced from Medline, Embase, and Cochrane databases. All randomised controlled comparisons of the effect of patient-specific instrumentation on the coronal alignment of total knee replacement were included. The main outcome was the risk difference measured by the proportion of failures in the control group minus the proportion of failures in the experimental group. Through Bayesian statistics, we estimated cumulatively over publication time of the trial results: the posterior probabilities that the risk difference was more than 5 and 10%; the posterior probabilities that given the results of all previous published trials an additional fictive trial would achieve a risk difference of at least 5%; and the predictive probabilities that observed failure rate differ from 5% across arms. RESULTS: Thirteen trials were identified including 1092 patients, 554 in the experimental group and 538 in the control group. The cumulative mean risk difference was 0.5% (95% CrI: -5.7%; +4.5%). The posterior probabilities that the risk difference be superior to 5 and 10% was less than 5% after trial #4 and trial #2 respectively. The predictive probability that the difference in failure rates was at least 5% dropped from 45% after the first trial down to 11% after the 13th. Last, only unrealistic trial design parameters could change the overall evidence accumulated to date. CONCLUSIONS: Bayesian probabilities are readily understandable when discussing the relevance of performing a new trial. It provides investigators the current probability that an experimental treatment be superior to a reference treatment. In case a trial is designed, it also provides the predictive probability that this new trial will reach the targeted risk difference in failure rates. TRIAL REGISTRATION: CRD42015024176 .

Learning curves for three specific procedures by anesthesiology residents using the learning curve cumulative sum (LC-CUSUM) test.

BACKGROUND: The learning curve cumulative sum (LC-CUSUM) test is an innovative tool that allows quantitative monitoring of individual medical performance during the learning process by determining when a predefined acceptable level of performance is reached. This study used the LC-CUSUM test to monitor the learning process and failure rate of anesthesia residents training for specific subspecialty anesthesia procedures. METHODS: The study included 490 tracheal punctures (TP) for jet ventilation, 340 thoracic epidural analgesia (TEA) procedures, and 246 fiberoptic nasal intubations (FONI) performed by 18 residents during their single 6-month rotation. RESULTS: Overall, 27 (14-52), 19 (5-41), and 14 (6-33) TP, TEA, and FONI procedures were performed, respectively, by each resident. In total, 2 of 18 residents achieved an acceptable failure rate for TEA according to the literature and 4 of 18 achieved an acceptable failure rate for FONI, while none of the residents attained an acceptable rate for TP. CONCLUSIONS: A single 6-month rotation in a reference teaching center may not be sufficient to train residents to perform specific or sub-specialty procedures as required. A regional learning network may be useful. More patient-based data are necessary to conduct a risk adjustment analysis for such specific procedures.

Validity and Reliability of Intraoperative Radiographs to Assess Leg Length During Total Hip Arthroplasty: Correlation and Reproducibility of Anatomic Distances.

BACKGROUND: Leg length discrepancy after total hip arthroplasty is a frequent complication. The aim of this study was to assess the validity (correlation) and reproducibility (inter-rater agreement) of various intraoperative hip radiographs measures to estimate leg length. METHODS: Patients were included if they were aged 15 years or older; were eligible for a total hip arthroplasty, and were operated in lateral recumbent. An intraoperative hip radiograph was performed with the definitive implants in place. At 6 weeks postoperatively, anteroposterior pelvis radiograph was taken. We used 3 measures to assess leg length: the height from the ischial tuberosity to the lesser trochanter (LTI), the height from the center of femoral head to the greater trochanter (GTC), and to the inferior teardrop (TC). RESULTS: The study group consisted of 71 hips with an average age of 69 years (range, 24-92 years). The correlation was 0.545 (95% CI: 0.35-0.69) for GTC, 0.75 (95% CI: 0.61-0.84) for TC, and 0.70 (95% CI: 0.56-0.80) for LTI. Intraoperative and postoperative measures were statistically different for GTC (<0.0001) and TC (<0.0001), and not significant for LTI (P = .06). Reproducibility of these measures were excellent with intraclass correlation coefficients of 0.977, 0.814, and 0.983 for the GTC, TC, and LTI, respectively. CONCLUSION: None of the parameters used to assess leg-length based on an intraoperative radiograph showed good correlation with the postoperative radiograph.

Proficiency in oocyte retrieval assessed by the learning curve cumulative summation test.

The number of procedures required for a trainee to reach proficiency in oocyte retrieval and the criteria applied to define performance are not well defined. To evaluate the learning curve of oocyte retrieval, this study prospectively evaluated three trainees over 6 months. Oocyte retrieval was monitored by the learning curve-cumulative summation test (LC-CUSUM), a specific statistical tool designed to indicate when a predefined level of performance is reached. Oocytes were retrieved from one ovary by the trainee and from the second ovary by a senior operator in a randomized manner. The main outcome measure was the ratio of oocytes collected and follicles aspirated. A trainee's ratio of ≥ 80% of the senior operator's defined success. From 17 to >50 procedures were necessary for the trainees to reach the predefined level of performance. Cumulative summation tests implemented after the learning phase confirmed that performance was maintained. The present study confirms the large variability in acquiring proficiency for surgical procedures. It provides an exportable model for a quantitative tailored monitoring of the learning curve and for continuous monitoring of performance in oocyte retrieval.

Student attainment of proficiency in a clinical skill: the assessment of individual learning curves.

The aims of this study were to determine if the learning curve cumulative summation test (LC-CUSUM) can differentiate proficiency in placing intravenous catheters by novice learners, and identify the cause of failure when it occurred. In a prospective, observational study design 6 undergraduate students with no previous experience of placing intravenous catheters received standardized training by a board certified veterinary anesthesiologist in intravenous catheter placement technique. Immediately following training, each student attempted 60 intravenous catheterizations in a dog mannequin thoracic limb model. Results were scored as a success or failure based upon completion of four specific criteria, and where catheter placement failure occurred, the cause was recorded according to pre-defined criteria. Initial acceptable and unacceptable failure rates were set by the study team and the LC-CUSUM was used to generate a learning curve for each student. Using 10% and 25% acceptable and unacceptable failure rates, 3 out of 6 students attained proficiency, requiring between 26 to 48 attempts. Applying 25% and 50% acceptable and unacceptable failure rates, 5 of 6 students obtained proficiency, requiring between 18 and 55 attempts. Wide inter-individual variability was observed and the majority of failed catheterisation attempts were limited to two of the four pre-defined criteria. These data indicate that the LC-CUSUM can be used to generate individual learning curves, inter-individual variability in catheter placement ability is wide, and that specific steps in catheter placement are responsible for the majority of failures. These findings may have profound implications for how we teach and assess technical skills.

Tumor endoprosthesis revision rates increase with peri-operative chemotherapy but are reduced with the use of cemented implant fixation.

Chemotherapy may reduce osseointegration of tumor endoprosthesis, while delaying chemotherapy may reduce survival. We studied the effects of chemotherapy and cemented fixation on tumor endoprosthesis survivorship with a retrospective analysis of 50 consecutive patients receiving lower limb salvage surgery. We compared rates of radiographic loosening/revision and effect of cement fixation between chemotherapy/no chemotherapy cohorts. Chemotherapy increased the total revision rate (HR = 3.8 [1-14], P = 0.033), but did not affect aseptic loosening. Cement fixation reduced revision for loosening (HR = 0.09 (0.008-0.98), P = 0.012) and showed less radiographic loosening (HR = 0.09 (0.02-0.51), P = 0.00066). Cement fixation had lower rates of revision for loosening and radiographic loosening regardless of whether chemotherapy was given. We conclude that for these implants, cement fixation provides superior results to uncemented fixation.

Learning curve of acetabular cup positioning in total hip arthroplasty using a cumulative summation test for learning curve (LC-CUSUM).

Despite advances in surgical techniques and instrumentation, optimal cup positioning in total hip arthroplasty (THA) is challenging with a limited accuracy. We evaluated whether a learning curve exists for the optimal cup positioning, using the LC-CUSUM test (Learning curve cumulative summation test). We evaluated the first 100 consecutive THA performed by two surgeons, who had a year of fellowship training in the same teaching hospital. A learning curve of cup positioning was plotted in each series using the LC-CUSUM score. There was no significant difference of numbers of outlier between two surgeons (P = 0.079). Both surgeons completed the learning curve of optimal cup positioning before 50 procedures, and maintained competence. A substantial learning period is necessary in the optimal positioning of an acetabular cup.

Monitoring the adequacy of surgical margins after resection of bone and soft-tissue sarcoma.

PURPOSE: Local recurrence of a bone or soft-tissue sarcoma is a devastating complication. Minimizing the proportion of positive surgical margins, or tumor contamination, during resection is of paramount importance. METHODS: Resections of sarcomas were prospectively evaluated and considered inadequate if unplanned microscopic or macroscopic positive surgical margins were identified or if inadvertent tumor contamination of the wound occurred. Monitoring of performance was continuously performed with a statistical process control method, the cumulative sum test, and regular meetings were held to discuss the reasons for failures. A target performance of 5 % inadequate procedures was chosen. RESULTS: A total of 146 sarcomas-106 soft tissue and 40 bone-were resected during the monitoring period. Six (4 %) procedures were considered inadequate: three patients had inadvertent tumor contamination of the wound, two patients had unplanned microscopic positive margins, and one patient had both. Performance was considered to be adequate during the whole monitoring period. CONCLUSIONS: With adequate preoperative planning and surgical technique, the risk of an inadequate resection can be limited. Implementation of a statistical process control method allows for ongoing performance monitoring and ensures that quality remains adequate over time.

Using the CUSUM test to control the proportion of inadequate open biopsies of musculoskeletal tumors.

BACKGROUND: Biopsies of musculoskeletal tumors lead to alterations in treatment in almost 20% of cases. Control charts are useful to ensure that a process is operating at a predetermined level of performance, although their use has not been demonstrated in assessing the adequacy of musculoskeletal biopsies. QUESTIONS/PURPOSES: We therefore (1) assessed the incidence of and the reasons for inadequate musculoskeletal biopsies when following guidelines for performing the procedure; and (2) implemented a process control chart, the CUSUM test, to monitor the proportion of inadequate biopsies. METHODS: We prospectively studied 116 incisional biopsies. The biopsy was performed according to 10 rules to (1) minimize contamination in the tissues surrounding the tumor; and (2) improve accuracy. A frozen section was systematically performed to confirm that a representative specimen was obtained. Procedures were considered inadequate if: (1) another biopsy was necessary; (2) the biopsy tract was not appropriately placed; and (3) the treatment provided based on the diagnosis from the biopsy was not appropriate. RESULTS: Five (4.3%) of the 116 incisional biopsy procedures were considered failures. Three patients required a second repeat open biopsy and two were considered to receive inappropriate treatment. No alarm was raised by the control chart and the performance was deemed adequate over the monitoring period. CONCLUSIONS: The proportion of inadequate musculoskeletal open biopsies performed at a referral center was low. Using a statistical process control method to monitor the failures provided a continuous measure of the performance.

No difference in quality-of-life outcomes after mobile and fixed-bearing medial unicompartmental knee replacement.

This is a retrospective study comparing the clinical outcomes of mobile vs fixed medial unicompartmental knee replacements. Patients were identified from a prospectively collected database. Demographic data and pre- and postoperative outcomes (Oxford, Western Ontario and McMaster Universities Osteoarthritis Index, Short Form Health Survey-12, and satisfaction) were collected-104 knees (90 patients) with minimum 2-years of follow-up: 37 mobile and 67 fixed-bearing. At baseline, the median age was 60 years (mobile) and 66 (fixed); body mass index was significantly higher (32) for the mobile group than for the fixed (28). At follow-up, the groups differed only in the Oxford score: median 83 (mobile) and 90 (fixed). There is no difference in outcomes between mobile and fixed unicompartmental knee replacements.

Effects of neoadjuvant chemotherapy on image-directed planning of surgical resection for distal femoral osteosarcoma.

BACKGROUND: Standard therapy for localized osteosarcoma includes neoadjuvant chemotherapy preceding local control surgery, followed by adjuvant chemotherapy. When limb-salvage procedures were being developed, preoperative chemotherapy allowed a delay in definitive surgery to permit fabrication of custom endoprosthetic reconstruction implants. One rationale for its continuation as the care standard has been the perception that it renders surgery easier and safer. Our objective was to compare surgical procedures planned on the basis of magnetic resonance images (MRIs) of distal femoral osteosarcomas acquired before neoadjuvant chemotherapy with surgical procedures planned on the basis of MRIs acquired after neoadjuvant chemotherapy as a measure of the surgically critical anatomic effects of the chemotherapy. METHODS: Twenty-four consecutive patients with distal femoral osteosarcoma had available digital MRIs preceding and following neoadjuvant chemotherapy. Thorough questionnaires were used to catalogue surgically critical anatomic details of MRI-directed surgical planning. Four faculty musculoskeletal oncologic surgeons and two musculoskeletal radiologists evaluated the blinded and randomly ordered MRIs. Interrater and intrarater reliabilities were calculated with intraclass correlation coefficients. The Student t test and chi-square test were used to compare pre-chemotherapy and post-chemotherapy continuous and categorical variables on the questionnaire. Mixed-effect regression models were employed to compare surgical procedures planned on the basis of pre-chemotherapy MRIs and with those planned on the basis of post-chemotherapy MRIs. RESULTS: The blinded reviews generated strong intraclass correlation coefficients for both interrater (0.772) and mean intrarater (0.778) reliability. The MRI-planned resections for the majority of tumors changed meaningfully after chemotherapy, but in inconsistent directions. On the basis of mixed-effect regression modeling, it appeared that more amputations were planned on the basis of post-chemotherapy MRIs. No other parameters differed in a significant and clinically meaningful fashion. Surgeons demonstrated their expectation that neoadjuvant chemotherapy would improve resectability by planning more radical surgical procedures on the basis of scans that they predicted had been obtained pre-chemotherapy. CONCLUSIONS: Surgeons can reliably record the anatomic details of a planned resection of an osteosarcoma. Such methods may be useful in future multi-institutional clinical trials or registries. The common belief that neoadjuvant chemotherapy increases the resectability of extremity osteosarcomas remains anecdotally based. Rigorous assessment of this phenomenon in larger cohorts and at other anatomic sites as well as re-evaluation of other arguments for neoadjuvant chemotherapy should be considered.

Local recurrence of localized soft tissue sarcoma: a new look at old predictors.

BACKGROUND: The objective of this study was to examine the effect of known predictors of local recurrence of soft tissue sarcoma in a competing risk setting. METHODS: The outcome of interest was the cumulative probability of local recurrence per category of relevant predictors, with death as a competing event. In total, 1668 patients with a localized soft tissue sarcoma of the extremity or trunk were included. RESULTS: Tumor size (hazard ratio, 3.3), depth (hazard ratio, 3.2), and histologic grade (hazard ratio, 4.5) were the variables that had the most effect on the risk of metastasis and, accordingly, were the most likely to induce competition. Surgical margins (hazard ratio, 3.3), histologic grade (hazard ratio, 2.1), presentation status (hazard ratio, 2.4), and tumor depth (hazard ratio, 1.5) were the variables that had the most effect on the risk of local recurrence. The 10-year cumulative probabilities of local recurrence were markedly different within categories for presentation status (P < .001) and surgical margin status (P < .001). However, because of the competing effect of death, there was little difference in the 10-year cumulative probabilities of local recurrence with regard to tumor depth (12% and 11.4% for deep and superficial tumors, respectively; P = .2), tumor size (10.6% and 13.3% for large and small tumors, respectively; P = .99), or histologic tumor grade (12.6%, 10.7%, and 11.1% for high, intermediate, and low-grade tumors, respectively; P = .17). CONCLUSIONS: Because of the competition between local recurrence and death, histologic tumor grade, tumor size, and tumor depth had little influence on the cumulative probability of local recurrence. The authors concluded that local management should be based on presentation status and surgical margins rather than other, previously acknowledged factors.

Monitoring the one year postoperative infection rate after primary total hip replacement.

PURPOSE: Infection of a total hip replacement is potentially a devastating complication. Statistical process control methods have been generating interest as a means of improving the quality of healthcare, and we report our experience with the implementation of such a method to monitor the one year infection rate after primary total hip replacement. METHOD: Infection was defined as the growth of the same organism in cultures of at least two aspirates or intra-operative specimens, or growth of one pathogen in a patient with local signs of infection such as erythema, abscess or draining sinus tract. The cumulative summation test (CUSUM test) was used to continuously monitor the one year postoperative infection rate. The target performance was 0.5% and the test was set to detect twice that rate. RESULTS: Over the three year study period, 2006 primary total hip replacements were performed. Infection developed within one year after surgery in eight (0.4%) hips. The CUSUM test generated no alarms during the study period, indicating that there was no evidence that the process was out of control. CONCLUSION: The one year infection rate after primary total hip replacement was in control. The CUSUM test is a useful method to continuously ensure that performance is maintained at an adequate level.

Axial Skeletal Location Predicts Poor Outcome in Ewing's Sarcoma: A Single Institution Experience.

Introduction. Ewing's sarcomas (EWSs) of bone and soft tissue are neuroectodermal tumors that affect both axial and appendicular locations. We hypothesized that axial location predicted poor outcome in EWS patients. Materials and Methods. Sixty-seven patients (57 with bone EWS and 10 with soft tissue EWS) were identified from our database. Thirty-four (51%) had axial EWS and 33 (49%) had appendicular EWS. Statistical analyses identified predictors of poor outcome. Results and Discussion. Axial location, large size, metastases at presentation, lack of definitive treatment, and positive surgical margins all correlated with poor outcome in univariate analysis. In multivariate analysis, axial location still predicted poor outcome when adjusted for pretreatment variables. Axial location was not statistically predictive of poor outcome when adjusted for treatment variables. Conclusions. Anatomic location has a negative effect on outcome in EWS that cannot be completely explained by pretreatment or treatment factors. Additional studies are required to determine if there is a biologic difference between axial and appendicular EWS.

Incidence, predictive factors, and prognosis of chondrosarcoma in patients with Ollier disease and Maffucci syndrome: an international multicenter study of 161 patients.

BACKGROUND: Enchondromatosis is characterized by the presence of multiple benign cartilage lesions in bone. While Ollier disease is typified by multiple enchondromas, in Maffucci syndrome these are associated with hemangiomas. Studies evaluating the predictive value of clinical symptoms for development of secondary chondrosarcoma and prognosis are lacking. This multi-institute study evaluates the clinical characteristics of patients, to get better insight on behavior and prognosis of these diseases. METHOD: A retrospective study was conducted using clinical data of 144 Ollier and 17 Maffucci patients from 13 European centers and one national databank supplied by members of the European Musculoskeletal Oncology Society. RESULTS: Patients had multiple enchondromas in the hands and feet only (group I, 18%), in long bones including scapula and pelvis only (group II, 39%), and in both small and long/flat bones (group III, 43%), respectively. The overall incidence of chondrosarcoma thus far is 40%. In group I, only 4 patients (15%) developed chondrosarcoma, in contrast to 27 patients (43%) in group II and 26 patients (46%) in group III, respectively. The risk of developing chondrosarcoma is increased when enchondromas are located in the pelvis (odds ratio, 3.8; p = 0.00l). CONCLUSIONS: Overall incidence of development of chondrosarcoma is 40%, but may, due to age-dependency, increase when considered as a lifelong risk. Patients with enchondromas located in long bones or axial skeleton, especially the pelvis, have a seriously increased risk of developing chondrosarcoma, and are identified as the population that needs regular screening on early detection of malignant transformation.