Evaluation of the EQ-5D-5L, EQ-VAS stand-alone component and Oxford knee score in the Australian knee arthroplasty population utilising minimally important difference, concurrent validity, predictive validity and responsiveness.

PURPOSE: To evaluate the Oxford Knee Score (OKS), EQ-5D-5L utility index and EQ-5D visual analogue scale (EQ-VAS) for health-related quality of life outcome measurement in patients undergoing elective total knee arthroplasty (TKA) surgery. METHODS: In this prospective multi-centre study, the OKS and EQ-5D-5L index scores were collected preoperatively, six weeks (6w) and six months (6 m) following TKA. The OKS, EQ-VAS and EQ-5D-5L index were evaluated for minimally important difference (MID), concurrent validity, predictive validity (Spearman's Rho of predicted and observed values from a generalised linear regression model (GLM)), responsiveness (effect size (ES) and standard response mean (SRM)). The MID for the individual patient was determined utilising two approaches; distribution-based and anchor-based. RESULTS: 533 patients were analysed. The EQ-5D-5L utility index showed good concurrent validity with the OKS (r = 0.72 preoperatively, 0.65 at 6w and 0.69 at 6 m). Predictive validity for the EQ-5D-5L index was lower than OKS when regressed. Responsiveness was large for all fields at 6w for the EQ-5D-5L and OKS (EQ-5D-5L ES 0.87, SRM 0.84; OKS ES 1.35, SRM 1.05) and 6 m (EQ-5D-5L index ES 1.31, SRM 0.95; OKS ES 1.69, SRM 1.59). The EQ-VAS returned poorer results, at 6w an ES of 0.37 (small) and SRM of 0.36 (small). At 6 m, the EQ-VAS had an ES of 0.59 (moderate) and SRM of 0.47 (small). It, however, had similar predictive validity to the OKS, and better than the EQ-5D-5L index. MID determined using anchor approach, was shown that for OKS at 6 weeks it was 8.84 ± 9.28 and at 6 months 13.37 ± 9.89. For the EQ-5D-5L index at 6 weeks MID was 0.23 ± 0.39, and at 6 months 0.26 ± 0.36. CONCLUSIONS: The EQ-5D-5L index score and the OKS demonstrate good concurrent validity. The EQ-5D-5L index demonstrated lower predictive validity at 6w, and 6 m than the OKS, and both PROMs had adequate responsiveness. The EQ-VAS had poorer responsiveness but better predictive validity than the EQ-5D-5L index. This article includes MID estimates for the Australian knee arthroplasty population.

A multi-disciplinary program for opioid sparse arthroplasty results in reduced long-term opioid consumption: a four year prospective study.

INTRODUCTION: The current opioid epidemic poses patient safety and economic burdens to healthcare systems worldwide. Postoperative prescriptions of opioids contribute, with reported opioid prescription rates following arthroplasty as high as 89%. In this multi-centre prospective study, an opioid sparing protocol was implemented for patients undergoing knee or hip arthroplasty. The primary outcome is to report our patient outcomes in the context of this protocol, and to examine the rate of opioid prescription on discharge from our hospitals following joint arthroplasty surgery. This is possibly associated with the efficacy of the newly implemented Arthroplasty Patient Care Protocol. METHODS: Over three years, patients underwent perioperative education with the expectation to be opioid-free after surgery. Intraoperative regional analgesia, early postoperative mobilisation and multimodal analgesia were mandatory. Long-term opioid medication use was monitored and PROMs (Oxford Knee/Hip Score (OKS/OHS), EQ-5D-5 L) were evaluated pre-operatively, and at 6 weeks, 6 months and 1 year postoperatively. Primary and secondary outcomes were opiate use and PROMs at different time points. RESULTS: A total of 1,444 patients participated. Two (0.2%) knee patients used opioids to one year. Zero hip patients used opioids postoperatively at any time point after six weeks (p < 0.0001). The OKS and EQ-5D-5 L both improved for knee patients from 16 (12-22) pre-operatively to 35 (27-43) at 1 year postoperatively, and 70 (60-80) preoperatively to 80 (70-90) at 1 year postoperatively (p < 0.0001). The OHS and EQ-5D-5 L both improved for hip patients from 12 (8-19) preoperatively to 44 (36-47) at 1 year postoperatively, and 65 (50-75) preoperatively to 85 (75-90) at 1 year postoperatively (p < 0.0001). Satisfaction improved between all pre- and postoperative time points for both knee and hip patients (p < 0.0001). CONCLUSIONS: Knee and hip arthroplasty patients receiving a peri-operative education program can effectively and satisfactorily be managed without long-term opioids when coupled with multimodal perioperative management, making this a valuable approach to reduce chronic opioid use.

Short-term difference only in reported outcomes (PROMs) after anterior or posterior approach to total hip arthroplasty: a 4-year prospective multi-centre observational study.

BACKGROUND: The direct anterior approach (DAA) in total hip arthroplasty (THA) may demonstrate better functional recovery compared to the posterior approach (PA). METHODS: In this prospective multi-centre study, patient-related outcome measures (PROMs) and length of stay (LOS) were compared between DAA and PA THA patients. The Oxford Hip Score (OHS), EQ-5D-5L, pain and satisfaction scores were collected at four perioperative stages. RESULTS: 337 DAA and 187 PA THAs were included. The OHS PROM was significantly better in the DAA group at 6 weeks post-operatively (OHS: 33 vs. 30, p = 0.02, EQ-5D-5L: 80 vs. 75, p = 0.03), but there were no differences at 6 months and at 1 year. EQ-5D-5L scores were similar between both groups at all time points. LOS as inpatient was significantly different, in favour of DAA [median 2 days (IQR 2-3) vs. PA 3 (IQR 2-4), p ≤ 0.0001]. CONCLUSIONS: Patients undergoing DAA THA have shorter LOS and report better short-term Oxford Hip Score PROMs at 6 weeks, but DAA did not convey long-term benefits over PA THA.

Association of anesthesia and analgesia with long-term mortality after hip fracture surgery: an analysis of the Australian and New Zealand hip fracture registry.

INTRODUCTION: Hip fractures are a common frailty injury affecting a vulnerable geriatric population. It is debated if anesthetic and analgesic techniques are associated with altered risk for outcomes in hip fracture patients. This study aimed to determine the association of anesthesia and regional analgesia with all cause 12-month mortality and even longer-term mortality after hip fracture surgery in Australia and New Zealand. METHODS: Data from the Australian and New Zealand Hip Fracture Registry collected from 2016 to 2018, with a minimum follow-up of 12 months, were reviewed. Anesthesia type and use of regional nerve blocks were investigated. The primary outcome was all cause 12-month mortality. RESULTS: 12-month mortality was 30.6% (n=5410) in a total of 17,635 patients. There was no difference in 12-month mortality between patients who received spinal or general anesthesia (p=0.238). The administration of a combination of general and spinal anesthesia for surgery to repair the fracture was an independent predictor of higher 12-month mortality (unadjusted complete case HR=1.17 (95% CI 1.04 to 1.31); p<0.001). Nerve blocks performed in both the emergency department (ED) and the operating theater (OT) were associated with reduced long-term mortality (median follow-up 21 months) with an unimputed unadjusted HR=0.86 (95% CI 0.77 to 0.96; p=0.043). CONCLUSION: There was no difference in the association of 12-month mortality between general and spinal anesthesia in patients undergoing hip fracture surgery. However, there was an association with a higher risk of 12-month mortality in patients who received both general and spinal anesthesia for the same surgery. Patients who received a regional nerve block in both the ED and the OT had a lower association of 12-month and longer-term mortality risk. The reasons for these findings remain unknown and should be the subject of further research investigation.

The Pericapsular Nerve Group (PENG) block combined with Local Infiltration Analgesia (LIA) compared to placebo and LIA in hip arthroplasty surgery: a multi-center double-blinded randomized-controlled trial.

BACKGROUND: The PEricapsular Nerve Group (PENG) block is a novel regional analgesia technique that provides improved analgesia in patients undergoing hip surgery while preserving motor function. In this study the PENG block was investigated for analgesia in elective total hip arthroplasty (THA). METHODS: In this multi-centre double-blinded randomized-controlled trial, in addition to spinal anesthesia and local infiltration analgesia (LIA), THA patients received either a PENG block or a sham block. The primary outcome was pain score (numeric rating scale 0-10) 3 h postoperatively (Day 0). Secondary outcomes were postoperative quadriceps muscle strength, postoperative Day 1 pain scores, opiate use, complications, length of hospital stay, and patient-reported outcome measures. RESULTS: Sixty patients were randomized and equally allocated between groups. Baseline demographics were similar. Postoperative Day 0, the PENG group experienced less pain compared to the sham group (PENG: 14 (47%) patients no pain, 14 (47%) mild pain, 2 (6%) moderate/severe pain versus sham: 6 (20%) no pain, 14 (47%) mild pain, 10 (33%) moderate/severe pain; p = 0.03). There was no difference in quadriceps muscle strength between groups on Day 0 (PENG: 23 (77%) intact versus sham: 24 (80%) intact; p = 0.24) and there were no differences in other secondary outcomes. CONCLUSIONS: Patients receiving a PENG block for analgesia in elective THA experience less postoperative pain on Day 0 with preservation of quadriceps muscle strength. Despite these short-term benefits, no quality of recovery or longer lasting postoperative effects were detected.

A longitudinal validation of the EQ-5D-5L and EQ-VAS stand-alone component utilising the Oxford Hip Score in the Australian hip arthroplasty population.

PURPOSE: To evaluate the measurement properties of the Oxford Hip Score (OHS), EQ-5D-5L utility index and EQ-5D-5L visual analogue scale (EQ-VAS) in patients undergoing elective total hip arthroplasty in Australia. METHODS: In this prospective multi-centre study, the OHS and EQ-5D-5L were collected preoperatively, six weeks (6w) and six months (6m) postoperatively. The OHS, EQ-VAS and EQ-5D-5L index were evaluated for concurrent validity, predictive validity (Spearman's Rho of predicted and observed values from a generalised linear regression model (GLM)), and responsiveness (effect size (ES) and standard response mean (SRM)). RESULTS: 362 patients were included in this analysis for 6w and 269 for 6m. The EQ-5D-5L index showed good concurrent validity with the OHS (r = 0.71 preoperatively, 0.61 at 6w and 0.59 at 6m). Predictive validity for EQ-5D-5L index was similar to OHS when regressed (GLM). Responsiveness was good at 6w (EQ-5D-5L index ES 1.53, SRM 1.40; OHS ES 2.16, SRM 1.51) and 6m (EQ-5D-5L index ES 1.88, SRM 1.70; OHS ES 3.12, SRM 2.24). The EQ-VAS returned poorer results, at 6w an ES of 0.75 (moderate) and SRM 0.8. At 6m the EQ-VAS had an ES of 0.92 and SRM of 1.00. It, however, had greater predictive validity. CONCLUSIONS: The EQ-5D-5L index and the OHS demonstrate strong concurrent validity. The EQ-5D-5L index demonstrated similar predictive validity at 6w and 6m, and both PROMs had adequate responsiveness. The EQ-VAS should be used routinely together with the EQ-5D-5L index. The EQ-5D-5L is suitable to quantify health-related quality of life in Australian hip arthroplasty patients.

Pericapsular nerve group block results in a longer analgesic effect and shorter time to discharge than femoral nerve block in patients after hip fracture surgery: a single-center double-blinded randomized trial.

OBJECTIVE: The pericapsular nerve group (PENG) block is a regional block that possibly provides better analgesia than that of the femoral nerve block (FNB) for hip fracture surgery. A randomized comparative trial performed in our institution showed that the PENG block may provide improved pain reduction compared with the FNB while preserving quadriceps strength. METHODS: In this single-center, double-blinded, randomized comparative trial, patients who underwent hip fracture surgery were randomized to receive either a FNB or PENG block for analgesia. This analysis reviews the outcomes of the block effect duration and time to discharge readiness. RESULTS: Sixty patients with similar baseline demographics were randomized. The median FNB duration was 15 hours, 35 minutes (range (hours:minutes) 4:08-30:45), and the median PENG duration was 22 hours, 50 minutes (range 6:00-32:00). The time to discharge readiness was shorter in the PENG group (3 days, range 1-14 days) than that in the FNB group (4 days, range 2-15 days). CONCLUSIONS: The PENG block results in a faster recovery and shorter time to discharge readiness. The duration of the PENG block appears to be longer than that of the FNB.

Pericapsular nerve group (PENG) block provides improved short-term analgesia compared with the femoral nerve block in hip fracture surgery: a single-center double-blinded randomized comparative trial.

BACKGROUND: The femoral nerve block (FNB) may be used for analgesia in hip fracture surgery. The pericapsular nerve group (PENG) block is a novel regional technique and may provide better pain reduction while preserving motor function, but these blocks have not been directly compared. METHODS: In a single-center double-blinded randomized comparative trial, patients presenting for hip fracture surgery received analgesia with either FNB or PENG block. The primary outcome measure was pain scores (Numeric Rating Scale (NRS) 0 to 10). Secondary outcomes were postoperative quadriceps strength, opiate use, complications, length of hospital stay, and patient-reported outcomes. RESULTS: Sixty patients were randomized and equally allocated between groups. Baseline demographics were similar. Postoperatively in recovery (day 0), the PENG group experienced less pain compared with the FNB group. (In the PENG group, 63% experienced no pain, 27% mild pain, and 10% moderate to severe pain. In comparison, 30% of the FNB group reported no pain, 27% mild pain, and 36% moderate to severe pain; p=0.04). This was assessed using an 11-point Likert NRS. Quadriceps strength was better preserved in the PENG group in the recovery unit (assessed using Oxford muscle strength grading, 60% intact in the PENG group vs none intact in the FNB group; p<0.001) and on day 1 (90% intact vs 50%, respectively; p=0.004). There was no difference in other outcomes. CONCLUSIONS: Patients receiving a PENG block for intraoperative and postoperative analgesia during hip fracture surgery experience less postoperative pain in the recovery room with no difference detected by postoperative day 1. Quadriceps strength was better preserved with the PENG block. Despite the short-term analgesic benefit and improved quadriceps strength, there were no differences detected in the quality of recovery.

Analgesia and anesthesia using the pericapsular nerve group block in hip surgery and hip fracture: a scoping review.

INTRODUCTION: Pericapsular nerve group (PENG) block is a novel regional analgesia technique to reduce pain after hip surgery and hip fractures. This review was conducted to summarize current literature. METHODS: A scoping review was carried out using the Joanna Briggs Institute framework. All articles describing the use of PENG block as a regional analgesia and/or anesthesia technique for hip pain were considered eligible for inclusion. Ovid Medline, Embase, CINAHL, PubMed and Google Scholar were searched. Adult and pediatric studies were included. Excluded were articles not available in English language, not available in full-text, related to non-orthopedic indications such as soft tissue surgery, and pelvic or femoral shaft fractures. RESULTS: Database searches identified 345 articles, 20 of which could be included in the current review, with a combined patient number of 74. Included articles comprised case reports and case series only, describing 1 to 10 patients. In all studies, PENG block was described to provide sufficient analgesia or anesthesia. Transient motor side effects occurred only when the local anesthetic was deposited in an unintended location (n=2). CONCLUSIONS: Current evidence of using PENG block for hip surgery or hip pain is limited to case reports and case series only. PENG block is a promising regional analgesia technique as an alternative to other regional nerve blocks such as femoral nerve block or iliac fascia nerve block. Observational and experimental studies are required to determine the effectiveness, efficacy and safety of the PENG block.

Is It the Surgery or the Block? Incidence, Risk Factors, and Outcome of Nerve Injury following Upper Extremity Surgery.

Although numerous studies have addressed the topic of postoperative nerve injury, debate continues to exist on its exact incidence, risk factors, etiology, and functional outcome. The aim of this study is to investigate the incidence of nerve injury and to identify patient, anesthetic, and surgical factors pertaining to perioperative nerve injury. Also, long-term nerve injury outcomes were assessed in terms of functionality. METHODS: A total of 297 patients, scheduled for elective distal upper extremity surgery, were prospectively included. At various time points, patients were screened for new onset nerve injury by means of clinical examination and questionnaires (including the Quick Disabilities of the Arm, Shoulder and Hand functionality measure). RESULTS: New nerve injury was diagnosed in 14 patients [4.7% (95% CI, 2.8-7.8)], but no causative risk factors were identified. The exact origin of nerve injury is suspected to be surgical in 11 cases. At 4 years postoperatively, 5 of the 14 patients with nerve injury (36%) were still symptomatic and had reduced functionality relative to preoperative status. CONCLUSIONS: This study demonstrates an incidence of all cause nerve injury of 4.7%. No specific patient, anesthetic, or surgical risk factors are identified and, importantly, patients who received regional anesthesia are not at more risk of nerve injury than those who received general anesthesia. The exact origin of nerve injury is very difficult to determine, but is suspected to be caused by direct surgical trauma in most cases. Four years following the nerve injury, approximately 40% of the patients with new onset nerve injury have reduced functionality.

Regional anaesthesia is associated with less patient satisfaction compared to general anaesthesia following distal upper extremity surgery: a prospective double centred observational study.

BACKGROUND: Patient satisfaction is a well-established indicator to evaluate the quality of medical care and there is an increasing support for the use of patient-reported experience measures (PREMs) to evaluate satisfaction. To anesthetize the upper limb for surgery, both general and regional plexus anaesthesia are appropriate techniques. However, the best technique in the anaesthesiologist's perspective might not necessarily result in the highest patient satisfaction. The aim of this study is to investigate patient satisfaction following general and regional anaesthesia, and to identify areas where anaesthesiologists can focus on improving patient care. METHODS: Patients scheduled for elective distal upper extremity surgery under either general or regional plexus anaesthesia were prospectively included. On the first postoperative day, patient satisfaction and main reason for dissatisfaction with the anaesthesia technique were investigated during a telephone interview. RESULTS: Of the 243 patients included in the current study, 79.8% report being "fully satisfied" with their anaesthesia technique. 32.1% of the patients who received regional anaesthesia reported not feeling "fully satisfied". This figure is 5.5% following general anaesthesia. Main reason for dissatisfaction following regional anaesthesia are reported as "insufficient anaesthesia prior to surgery", and "the discomfort of having a long-lasting insensate extremity postoperatively". CONCLUSIONS: Following regional plexus anaesthesia, a third of the patients are not "fully satisfied". To optimize patient satisfaction following regional anaesthesia techniques, we advocate stronger focus on patient counselling preoperatively, addressing the issues of block failure and prolonged postoperative sensory and motor block.

Individual duration of axillary brachial plexus block is unpredictable: a prospective double centered observational study.

BACKGROUND: The duration of effect for axillary plexus block using ropivacaine is highly variable. The available literature does not offer any plausible means of predicting time of block offset for individual patients, making it difficult to give accurate information and plan postoperative analgesics. This study was designed to identify factors influencing axillary plexus block offset time. METHODS: A total of 92 patients participated in this prospective double centred observational study. All patients were scheduled for axillary plexus block with ropivacaine 0.75% and subsequent block duration was recorded. RESULTS: Mean time of axillary plexus block offset was 13.5 hours, with a range of 4.8 to 25.4 hours. No statistical significant differences in offset time was seen with regard to gender, age, body weight, BMI and ASA-classification. A trend for increasing duration of blocks associated with increasing age was observed. No statistically significant difference was identified in block duration between blocks performed with nerve stimulator guidance versus ultrasound guidance. Similarly, neither dose nor volume of ropivacaine 0.75% was identified as a factor influencing block duration. CONCLUSIONS: This prospective study demonstrates a large inter individual variation in time of axillary plexus block offset using ropivacaine 0.75%. The lack of association between offset time and both demographic and block performance factors, makes predictability of individual duration of axillary plexus blocks in clinical practice extremely difficult. We suggest that all patients should be made aware of such variability in duration prior to block placement.

Safety and efficacy of isolated limb infusion with cytotoxic drugs in elderly patients with advanced locoregional melanoma.

INTRODUCTION: The treatment of elderly patients with advanced metastatic melanoma confined to a limb remains controversial. Isolated limb infusion (ILI) is an effective minimally invasive alternative to isolated limb perfusion (ILP) and is therefore a potentially valuable therapeutic option for this group. METHODS: From our prospective database 185 patients with advanced metastatic melanoma of the limb treated with a single ILI between 1992 and 2007 were identified. In all patients a cytotoxic combination of melphalan and actinomycin-D was used. RESULTS: Eighty-six patients (46%) were >or=75 years of age (range: 75-93). The patient characteristics in both groups were comparable except that the older group comprised more women (71% vs. 54%; P = 0.02) and had a lower body mass index (median: 24.4 vs. 26.4; P = 0.008). Complete response rates were 34% for those >or=75 years and 41% in the younger group (P = 0.28). There was no difference in limb recurrence free interval after a complete response (median: 24 months for both groups; P = 0.51) or in survival (median: 36 months for <75, 39 months for >or=75; P = 0.36) between both groups. Older patients experienced less limb toxicity after the procedure (Wieberdink grade III/IV toxicity in 36%) compared with younger patients (51%; P = 0.009) while systemic toxicity, complications, and long-term morbidity were similar. CONCLUSIONS: Elderly patients with advanced metastatic melanoma of the limb experience the same or lower toxicity after ILI compared with younger patients while response rates, limb recurrence free interval, survival, and morbidity are similar. ILI is an attractive alternative to the more laborious ILP, especially for older patients.

Major amputation for irresectable extremity melanoma after failure of isolated limb infusion.

INTRODUCTION: Isolated limb infusion (ILI) is an effective, minimally invasive treatment option that delivers high-dose regional chemotherapy to treat metastatic melanoma confined to a limb. In some patients, however, locoregional disease does not respond to the treatment or extensive recurrence occurs so that an amputation may become inevitable. In this study we analyzed indications for and results of amputation in these cases. METHODS: 14 patients were identified in whom amputation of the affected limb had to be carried out after failure of ILI. RESULTS: Following ILI, three patients had a complete response, seven had a partial response, two had stable disease and two patients had progressive disease. The median duration of response after ILI was 7 months (range 2-30). The median interval between ILI and amputation was 10 months. Amputation was performed in six of 20 patients who had been treated with an upper limb ILI, compared to eight amputations that were performed in 215 patients who had been treated with a lower limb ILI (P = .001). The indications for amputation were severe pain due to progression of tumor (n = 3), uncontrollable and troublesome tumor progression (n = 6) and bleeding from ulcerated lesions (n = 5). Five patients developed stump recurrence after amputation; these were treated by excision or radiation. Six of the eight patients who had a lower limb amputation became ambulant with the aid of prosthesis. Median survival after amputation was 13 months: three patients survived more than 5 years. CONCLUSIONS: Amputation following upper extremity ILI is more common compared to lower extremity ILI. Amputation may provide effective long-term palliation in selected patients when there is extensive inoperable progressive or recurrent disease after ILI.

Efficacy of repeat isolated limb infusion with melphalan and actinomycin D for recurrent melanoma.

BACKGROUND: : Isolated limb infusion (ILI) is an effective and minimally invasive treatment option for delivering regional chemotherapy in patients with metastatic melanoma confined to a limb. Recurrent or progressive disease after an ILI, however, presents a challenge for further treatment. The value of repeat ILI in this situation has not been well documented. METHODS: : Forty-eight patients were identified who had been treated with a repeat ILI. In all patients, a cytotoxic combination of melphalan and actinomycin D was used. RESULTS: : The median time between the 2 procedures was 11 months. The complete response (CR) rate after repeat ILI was 23%, compared with 31% after the initial ILI (P = .36). The overall response was 83%, compared with 75% after the first procedure (P = .32). The median duration of response was 11 months (10 months for patients with CR; P = .80), and median survival was 38 months. In those patients achieving a CR, the median survival was 68 months (P = .003). Toxicity after repeat ILI was increased, with 20 patients experiencing Wieberdink grade III limb toxicity (considerable erythema and edema with blistering) and 5 patients experiencing grade IV toxicity (threatened or actual compartment syndrome), whereas after the initial ILI these toxicity grades occurred in 14 patients and 1 patient, respectively (P = .03). No patient experienced grade V toxicity (requiring amputation). CONCLUSIONS: : Repeat ILI is an attractive treatment option to achieve limb salvage in patients with inoperable recurrent or progressive melanoma after a previous ILI. It can be associated with significant short-term regional toxicity, but is well tolerated by most patients, with satisfactory response rates. Cancer 2009. (c) 2009 American Cancer Society.

Isolated limb infusion as palliative treatment for advanced limb disease in patients with AJCC stage IV melanoma.

INTRODUCTION: In the treatment of patients with advanced limb melanoma a major treatment dilemma can arise when distant metastases are present also. Isolated limb infusion (ILI) has proved to be a useful limb-saving treatment and could potentially be of palliative value in patients with American Joint Committee on Cancer (AJCC) stage IV melanoma. METHODS: We identified 37 patients with advanced symptomatic limb disease as well as documented distant metastases at the time of their ILI. In all patients a drug combination of melphalan and actinomycin D was used. RESULTS: Fifty one percent had visceral distant metastases and 49% had cutaneous distant metastases only. The overall response rate in the treated limb was 76% [complete response (CR) rate 22%, partial response (PR) rate 54%]. Median response duration was 11 months (28 months for patients with CR; p = 0.08). Median survival after CR was 22 months, 17 months after PR, and only 4 months for those with stable or progressive disease (p = 0.002). Patients with visceral distant metastases had a significantly decreased survival compared with those with cutaneous distant metastases only (8 and 21 months, respectively; p = 0.03). Limb salvage was achieved in 86% of the patients. The procedure was well tolerated, with only one patient developing Wieberdink grade IV toxicity (threatened/actual compartment syndrome) and none requiring amputation as a result of the procedure (grade V toxicity). CONCLUSIONS: Minimally invasive ILI can effectively be used as palliative treatment to provide local tumor control and limb salvage in stage IV melanoma patients with advanced, symptomatic limb disease.