Predicting factors of postoperative relapse in T2-4N0M0 colorectal cancer patients via harvesting a minimum of 12 lymph nodes

Background and aim  The aim of this retrospective study was to determine which clinicopathological factors influenced the incidence of postoperative relapse and overall survival rates after radical resection of T_2-4N₀M₀ colorectal cancer (CRC) patients via harvesting a minimum of 12 lymph nodes. Materials and methods  Between January 2001 and June 2006, a total of 342 T_2-4N₀M₀ CRC patients who underwent radical resection were retrospectively analyzed in Kaohsiung Medical University Hospital. Of these 342 patients, 155 were observed by harvesting a minimum of 12 lymph nodes. These 155 patients were followed up intensively, and their outcomes were investigated retrospectively. Results  Of 155 patients, 83 were men (53.5%) and 72 (46.5%) were women. The mean age was 65.5 ± 11.1 years (range, 24–89 years). The median follow-up period was 49 months (range, 19–80 months). The present data showed invasive depth (P = 0.012), vascular invasion (P < 0.001), and perineural invasion (P = 0.009) as significantly prognostic factors for postoperative 5-year relapse rate by Kaplan–Meier analysis. Likewise, invasive depth (P = 0.013), vascular invasion (P < 0.001), and perineural invasion (P = 0.008) were significant factors for postoperative 5-year survival rate. Meanwhile, using a Cox proportional hazards analysis, depth of tumor invasion (P = 0.026) and vascular invasion (P = 0.001) were the independent predictors for postoperative relapse. Furthermore, the presence of vascular invasion was considerably correlated to the higher postoperative relapse rate and the poorer overall survival rates by survival analyses (P < 0.0001). Conclusions  Besides the conventional depth of tumor invasion, this study highlights the potential for using vascular invasion as a means of identifying a subgroup of T_2-4N₀M₀ CRC patients with adequate lymph node harvest at higher risk who would potential benefit from adjuvant therapy after surgery.     

Major liver resections in the elderly—is an aggressive approach justified?

Background and aims  As the mean life expectancy rises, the incidence of patients 75 years of age and older who present with colorectal liver metastases continues to increase. The purpose of our study was to evaluate the outcome of major hepatic resections in the elderly population. Patient and methods  From April 1998 to December 2006, 572 consecutive patients with colorectal liver metastases were treated at our Institution. Of these, 59 were 75 years or older. There was an intent to proceed with major liver resections in all cases. Data were analyzed according to diagnosis, comorbidities, extent of liver resection, postoperative complications, overall survival, and disease-free survival. Results  Surgical treatment included right hepatectomies (n = 8), left hepatectomies (n = 4), and sectionectomies (more than three segments; n = 33). Fourteen (n = 14) patients received an explorative laparotomy alone. Morbidity and hospital mortality were 10% and 3%, respectively. Overall survival of 1, 3, and 5 years was 90%, 64%, and 33%, respectively. The corresponding disease-free survival was 74%, 42%, and 32%. Resection margin (R class) was the only predictor of survival by both uni- and multivariate analyses. Conclusion  Hepatic resections can be performed safely in selected patients 75 years of age or older.     

Microvessel density of hepatocellular carcinoma: its relationship with prognosis

Purpose: To elucidate the relationship between angiogenesis and prognosis after curative resection of hepatocellular carcinoma (HCC). Methods: An immunohistochemical study using anti-CD34 monoclonal antibody was carried out on surgical specimens from 78 HCC patients who had undergone curative resection; microvessel density (MVD) was counted and the overall survival and disease-free survival were analyzed retrospectively. Results: Blood vessels in the tumor were strongly stained by anti-CD34 antibody, but not those in the surrounding liver parenchyma. There were three types of tumor vessels: capillary-like (n = 59), sinusoid-like (n = 16) and mixed-type (n = 3). The median MVD count was 100 per field. The HCC were designated as hypovascular (n = 36) with an MVD count below 100, and hypervascular (n = 42) with an MVD count of 100 or more per field. The 5-year survival and disease-free survival rates were 49.7% and 42.8% respectively, and statistical analysis showed that the MVD level was not correlated with tumor size, capsule status, Edmondson's grade, α-fetoprotein level, associated cirrhosis, γ-glutamyltransferase, and serum HBsAg status. The sinusoid-like tumor vessels appeared more frequently in the more differentiated tumors (P < 0.05). No statistical difference in overall and disease-free survival between different MVD levels and microvessel types was found. Tumor size was the only predicting factor in the entire series. In patients with small HCC (≤ 5 cm, n = 40), 5-year survival and disease-free survival rates were 58.9% and 52.7% respectively, higher than the values in large HCC (39.8% and 32.0% respectively, P < 0.05). The MVD level was an independent predicting factor of disease-free survival, 5-year disease-free survival in the hypovascular group (74.6%) being better than that in the hypervascular group (34.7%, P < 0.05). Conclusions: The MVD level was not related to tumor size, capsule statuo, Edmondson's grade, α-fetoprotein level, associated cirrhosis, γ-glutamyltransferase and serum HBsAg status. In the entire series, tumor size was the only factor influencing survival after curative resection. However, in patients with small HCC, the MVD level was an independent factor of disease-free survival. The pathological and clinical implications of different types of tumor vessels in HCC remain to be studied. Received: 27 November 1998 / Accepted: 5 January 1999     

Preoperative serum levels of serum VEGF-C is associated with distant metastasis in colorectal cancer patients

Purpose  Vascular endothelial growth factor-C (VEGF-C) is one of the most potent lymphangiogenic members of the VEGF family that has been associated with lymph node metastasis and poor prognosis in patients with colorectal cancer (CRC). In this study, we evaluated the relationship of preoperative serum VEGF-C (sVEGF-C) and survival in CRC patients. Materials and methods  sVEGF-C levels were determined, prior to resection, in a cohort of 120 newly presenting patients with CRC by quantitative ELISA. Results  Patients who had positive lymph node involvement and higher Dukes’ staging (C&D) were associated with shorter time to metastases as expected (p = 0.002 and 0.001, respectively). Patients with distant metastasis had significantly lower levels of sVEGF-C than those without histopathologically proven disease (p = 0.004). However, there was no significant difference in the median sVEGF-C level in patients with or without lymph node metastatic involvement (91 pg/ml vs. 124 pg/ml; p = 0.81). Patients with a sVEGF-C concentration less than the median value (103 pg/ml) showed a poorer overall survival than patients with sVEGF-C levels greater than the median; but this was not statistically significant. Conclusions  In this study, low sVEGF-C levels are associated with distant metastasis; hence, preoperative levels may aid in the selection of CRC patients who require further investigation.     

Direct Tumor Invasion in Colon Cancer: Correlation with Tumor Spread and Survival

Purpose  This study examined the correlation between depth of local invasion in colon cancer and tumor spread and patient survival. Methods  A cohort of 796 patients with a complete set of TNM staging information following an elective resection for colon cancer was selected. The rates of lymph node and distant metastasis, tumor differentiation, and extramural venous invasion for different tumor (T) categories were compared. The effects of initial tumor (T) category on overall patient survival were studied. Results  The depth of local tumor invasion correlated strongly with nodal involvement (P = 0.0001), rates of extramural venous invasion (P = 0.0002), poor differentiation (P = 0.0001), and distant metastasis (P = 0.0001). Fifty-seven percent of the patients remained lymph node-negative and distant metastasis-negative irrespective of their depth of tumor invasion had no impact on overall survival (P = 0.49). For patients with lymph node or distant metastasis (43 percent), depth of tumor invasion had significant impact on overall survival (P = 0.001). Thirteen percent of T3N1, 33 percent of T3N2, 40 percent of T4N1, and 68.percent of T4N2 cases had distant metastasis at presentation. Conclusion  Two types of colon cancer were observed: locally active and tendency to metastasize. For the latter, overall mortality and the risk of metastasis increased with depth of tumor invasion. Reprints are not available.     

Risk Factors and Oncologic Impact of Anastomotic Leakage after Rectal Cancer Surgery

Purpose  The impact of anastomotic leakage on the long-term oncologic outcome is not clear. This retrospective study evaluated risk factors and oncologic impacts of anastomotic leakage after rectal cancer surgery. Methods  Data were analyzed from 1,391 patients who underwent sphincter preservation for rectal cancer between January 1997 and August 2003. Operations were classified as anterior resection (n = 164), low anterior resection (n = 898), or ultralow anterior resection (n = 329). Results  The anastomotic leakage rate was 2.5 percent. Multivariate analysis identified male (hazard ratio, 3.03), old age (hazard ratio, 2.42), and lower anastomosis level (hazard ratio, 2.68) as risk factors for leakage. The local recurrence rates were 9.6 and 2.2 percent for the leakage and nonleakage groups, respectively but were not significant (P = 0.14). The overall five-year survival rates were 55.1 and 74.1 percent in the leakage and nonleakage groups, respectively (P < 0.05), and the cancer-specific survival rates were 63 and 78.3 percent in the leakage and nonleakage groups, respectively (P = 0.05). However, in subgroup analysis, significant differences were identified only in Stage III patients. Conclusions  Age, sex, and ultralow anterior resection were found to be risk factors for anastomotic leakage after rectal cancer surgery. In addition, leakage was associated with poor survival. Poster presentation at the meeting of The American Society Colon and Rectal Surgeons, Seattle Washington, June 3 to 7, 2006.     

Determinants of morbidity and survival after elective non-curative resection of stage IV colon and rectal cancer

Purpose  The benefit of elective primary tumor resection for non-curable stage IV colorectal cancer (CRC) remains largely undefined. We wanted to identify risk factors for postoperative complications and short survival. Methods  Using a prospective database, we analyzed potential risk factors in 233 patients, who were electively operated for non-curable stage IV CRC between 1996 and 2002. Patients with recurrent tumors, resectable metastases, emergency operations, and non-resective surgery were excluded. Risk factors for increased postoperative morbidity and limited postoperative survival were identified by multivariate analyses. Results  Patients with colon cancer (CC = 156) and rectal cancer (RC = 77) were comparable with regard to age, sex, comorbidity, American Society of Anesthesiologists score, carcinoembryonic antigen levels, hepatic spread, tumor grade, resection margins, 30-day mortality (CC 5.1%, RC 3.9%) and postoperative chemotherapy. pT4 tumors, carcinomatosis, and non-anatomical resections were more common in colon cancer patients, whereas enterostomies (CC 1.3%, RC 67.5%, p < 0.0001), anastomotic leaks (CC 7.7%, RC 24.2%, p = 0.002), and total surgical complications (CC 19.9%, RC 40.3%, p = 0.001) were more frequent after rectal surgery. Independent determinants of an increased postoperative morbidity were primary rectal cancer, hepatic tumor load >50%, and comorbidity >1 organ. Prognostic factors for limited postoperative survival were hepatic tumor load >50%, pT4 tumors, lymphatic spread, R1–2 resection, and lack of chemotherapy. Conclusions  Palliative resection is associated with a particularly unfavorable outcome in rectal cancer patients presenting with a locally advanced tumor (pT4, expected R2 resection) or an extensive comorbidity, and in all CRC patients who show a hepatic tumor load >50%. For such patients, surgery might be contraindicated unless the tumor is immediately life-threatening. Financial support  Neither one of the authors nor the institutions from which the work originated have asked for, accepted, or received any direct or indirect financial support from a third party regarding the matter and materials discussed in this paper.     

Colorectal liver metastasis surgery: analysis of risk factors predicting postoperative complications in relation to the extent of resection

Background/aims  Despite advances in diagnosis and treatment, the rate of complications after resection for colorectal liver metastases remains high. An awareness of risk factors is essential for the rates of morbidity and mortality to fall to optimal levels. Materials and methods  Of the 240 patients who underwent resection for the first manifestation of colorectal liver metastases, 49 patients with lobectomy or extended hepatectomy (major resections) and 58 with wedge resections within only one liver segment (minor resections) form the basis of this report. A total of 16 variables were analyzed to find the risk factors linked to postoperative morbidity and mortality. Results/findings  Thirty-four patients (31.8%) suffered postoperative complications, and one patient died during the hospital stay (0.9%). In the major resection group, multivariate analysis showed that neoadjuvant chemotherapy [odds ratio (OR): 2.4; p = 0.005], vascular clamping (OR: 1.4; p = 0.008), and intraoperative blood loss with transfusion of three to six packed red cell units (OR: 1.2; p = 0.029) were significantly associated with postoperative morbidity. Vascular clamping was an independent predictor for biliary fistula (OR: 1.2; p = 0.029). Postoperative temporary liver failure was influenced by neoadjuvant chemotherapy (OR: 3.4; p = 0.010), vascular clamping (OR: 1.5; p = 0.015), and requirement of blood transfusion (OR: 2.1; p = 0.016). After minor resections, only a decreased postoperative serum cholinesterase B level was an independent predictor for complications (OR: 2.2; p = 0.001), as well as for hemorrhage (OR: 1.6; p = 0.023). Postoperative mortality was not predicted by any of the factors that were analyzed. Interpretation/conclusion  Factors for complications differ depending on the extent of colorectal liver metastasis resection. Only knowledge and particular consideration of these factors may provide for an optimal postoperative outcome for the individual patient. Ralf Konopke and Stephan Kersting contributed equally to this work     

Metabolic consequences of a 50% partial pancreatectomy in humans

Aims/hypothesis  Partial pancreatectomy is frequently performed in patients with pancreatic tumours or chronic pancreatitis, but little is known about the metabolic impact of this intervention. We examined the effects of approximately 50% partial pancreatectomy on glucose homeostasis and insulin secretion. Methods  Fourteen patients with chronic pancreatitis, ten patients with pancreatic carcinoma and 13 patients with benign pancreatic tumours or extra-pancreatic masses (control group) underwent 240 min oral glucose tolerance tests before and after pancreatic tail-resection (n = 12), duodenopancreatectomy (n = 19) or duodenum-preserving pancreatic-head resection (n = 6). Results  Partial pancreatectomy led to a reduction in post-challenge insulin excursions by 49% in chronic pancreatitis patients, 52% in carcinoma patients and 55% in controls (p < 0.05). Nevertheless, post-challenge glucose concentrations were transiently ameliorated after surgery (p < 0.001). In the control participants, pancreatic-head resection caused a transient reduction of post-challenge glycaemia, whereas pancreatic-tail resection increased both fasting and post-challenge glycaemia (p < 0.05). Insulin sensitivity was highest in chronic pancreatitis patients before surgery (p < 0.01), but remained unchanged by the partial pancreatectomy. High pre-operative body weight and elevated fasting glucose levels were associated with poor glycaemic control after surgery. Conclusions/interpretation  Insulin secretion is diminished after pancreatic-head and -tail resection, but post-challenge glucose concentrations can be ameliorated after pancreatic-head resection. These data highlight the unequal impact of different surgical procedures on glucose control and suggest that obesity and high pre-operative glucose levels should be considered as risk factors for the development of hyperglycaemia after pancreatic surgery. B. A. Menge and H. Schrader contributed equally to this study     

Pulmonary resection for metastases from colorectal cancer: prognostic factors and survival

Backgrounds Pulmonary metastases occur in up to 10% of all patients who undergo curative resection. Surgical resection is an important part in the treatment of pulmonary metastasis from colorectal cancer. We analyzed the treatment outcome and prognostic factors affecting survival in this subset of patients. Materials and methods Between October 1994 and December 2004, 59 patients underwent curative resection for pulmonary metastases of colorectal cancer. Uncontrollable synchronous liver and lung metastasis or synchronous colorectal cancer with isolated lung metastasis were excluded from this study. A retrospective review of patient characteristics and factors influencing survival was performed. Survival was analyzed by the Kaplan–Meier method. Comparison between groups were performed by a log-rank analysis and the Cox proportional hazard model. Results The 5-year overall survival rate of all patients who received pulmonary resection was 50.3%. The number of pulmonary metastases was significantly related with survival in univariate analysis, but not in multivariate analysis (p = 0.032). Prethoracotomy carcinoembryonic antigen (CEA) level exceeding 5 ng/ml was related with poor survival (p = 0.001). A disease-free interval of greater than 2 years did not correlate with survival after thoracotomy (p = 0.3). Conclusion The prethoracotomy CEA level and the number of metastases were independent prognostic factors. Resection of pulmonary metastasis from colorectal cancer may result in improved survival or even healing in selected patients. Pulmonary resection of colorectal cancer is regarded as a safe and effective treatment with low morbidity and mortality rates.     

Systemic cytokine response after emergency and elective surgery for colorectal carcinoma

Background  Systemic cytokines (SC) are accepted mediators of host immune response. It is debated if long-term survival is influenced by emergency presentation of colorectal cancer, and the role of immunitary response is still unknown. The aim of this prospective study was to compare the SC response after emergency resection with that after elective resections of colorectal carcinoma. Materials and methods  One hundred six consecutive subjects with colorectal cancer were submitted to emergency (complete bowel obstruction; EMS, n = 50) or elective resection (ELS, n = 56) of the tumour. Sera were collected before surgery and at appropriate time points afterward and assayed for interleukin-1beta (IL-1β), tumour necrosis factor-alpha (TNF-α), interleukin-6 (IL-6) and C-reactive protein (CRP). Five-year survival was analysed according to Kaplan–Meier test. The Cox proportional hazard model was used for the multivariate analysis. Results  Pre-operative levels of IL-1β, IL-6 and CRP were statistically higher in the EMS group. Levels of TNF-α were not elevated after surgery and there was no difference between the groups. Five-year survival was significantly lower in the EMS group (p < 0.05). Conclusions  Immunitary response, as reflected by SC, was better after elective resection than after emergency resection of colorectal carcinoma and this difference may have implication in the long-term survival.     

Autologous hematopoietic stem cell transplantation—what determines the outcome: an experience from North India

Limited information is available from developing countries about complications, pattern of infections, and long-term outcome of patients following high-dose chemotherapy (HDCT) and autologous blood stem cell transplantation (ASCT). Between April, 1990 and December 2009, 228 patients underwent ASCT. Patients’ median age was 48 years, ranging from 11 to 68 years. There were 158 males and 70 females. Indications for transplant included multiple myeloma, n = 143; lymphoma, n = 44 (Hodgkin’s, n = 25 and non-Hodgkin’s, n = 19); leukemia, n = 22; and solid tumors, n = 18. Patients received HDCT as per standard protocols. Following ASCT, 175 (76.7%) patients responded; complete, 98 (43%); very good partial response, 37 (16.2%); and partial response, 40 (17.5%). Response rate was higher for patients with good Eastern Cooperative Oncology Group (ECOG) performance status (0–2 vs. 3–4, p < 0.001), pretransplant chemo-sensitive disease (p < 0.001) and those with diagnosis of hematological malignancies (p < 0.003). Mucositis, gastrointestinal, renal, and liver dysfunctions were major nonhematologic toxicities, 3.1% of patients died of regimen-related toxicities. Infections accounted for 5.3% of deaths seen before day 30. At a median follow-up of 66 months (range, 9–234 months), median overall (OS) and event-free survival (EFS) were 72 months (95% CI 52.4–91.6) and 24 months (95% CI 17.15–30.9), respectively. For myeloma, OS and EFS were 79 months (95% CI 52.3–105.7) and 30 months (95% CI 22.6–37.4), respectively. Pretransplant good performance status and achievement of significant response following transplant were major predictors of survival. Our analysis demonstrates that such procedure can be successfully performed in a developing country with results comparable to developed countries.     

No prognostic impact of isolated lymphovascular invasion after radical resection of rectal cancer—results of a multicenter observational study

Background The prognostic impact of isolated lymphovascular invasion (LVI) after radical resection of rectal cancer is controversially discussed. However, it could be relevant to decide for an adjuvant treatment. Aim The aim of the analysis was, based on the data of an observational study, to determine the prognostic relevance of the isolated LVI. Materials and methods Patients after radical resection of rectal cancer with no hemangioinvasion were subdivided in three groups: I—no LVI, no lymph node metastases (LNM); II—positive LVI, no LNM; III—positive LNM. Five-year local recurrence rate, distant metastases-free and disease-free survival were determined uni- and multivariate. Results Patients, n = 846, were studied (I, n = 471; II, n = 75; III, n = 300). The univariate comparison between the groups revealed the following 5-year results: local recurrence rate: 9.4 vs 10.0 vs 14.0%; distant metastases-free survival: 84.1 vs 82.5 vs 49.3%; disease-free survival: 83.2 vs 80.7 vs 45.5%. The differences between groups I and III were significant, but not between groups I and II. The determined higher disease-free survival rate in group II vs group III was significant (P = 0.041), but the differences in local recurrence rate and rate of distant metastases did not reach statistical significance. The multivariate analysis revealed no impact of the isolated LVI on the oncological outcome. Conclusion The isolated LVI has no independent prognostic impact on the local recurrence rate and long-term survival after radical resection of rectal cancer. Based on this finding, no indication for an adjuvant treatment in these patients can be derived.     

Simultaneous versus staged liver resection of synchronous liver metastases from colorectal cancer

Background and aims The surgical strategy for treatment of synchronous liver metastases from colorectal cancer remains controversial. This retrospective analysis was conducted to compare the postoperative outcome and survival of patients receiving simultaneous resection of liver metastases and primary colorectal cancer to those receiving staged resection. Materials and methods Between January 1988 and September 2005, 219 patients underwent liver resection for synchronous colorectal liver metastases, of whom, 40 patients received simultaneous resection of liver metastases and primary colorectal cancer, and 179 patients staged resections. Patients were identified from a prospective database, and records were retrospectively reviewed. Patient, tumor, and operative parameters were analyzed for their influence on postoperative morbidity and mortality as well as on long-term survival. Results Simultaneous liver resections tend to be performed for colon primaries rather than for rectal cancer (p = 0.004) and used less extensive liver resections (p < 0.001). The postoperative morbidity was comparable between both groups, whereas the mortality was significantly higher in patients with simultaneous liver resection (p = 0.012). The mortality after simultaneous liver resection (n = 4) occurred after major hepatectomies, and three of these four patients were 70 years of age or older. There was no significant difference in long-term survival after formally curative simultaneous and staged liver resection. Conclusion Simultaneous liver and colorectal resection is as efficient as staged resections in the treatment of patients with colorectal cancer and synchronous liver metastases. To perform simultaneous resections safely a careful patient selection is necessary. The most important criteria to select patients for simultaneous liver resection are age of the patient and extent of liver resection.     

Twenty-year survivors after resection for hepatocellular carcinoma-analysis of 53 cases

Purpose  To clarify the clinicopathologic features of patients surviving ≥20 years after resection for hepatocellular carcinoma (HCC). Methods  Between 1961 and 1987, a total of 396 patients underwent hepatic resection for HCC; 53 (13.4%) patients survived ≥20 years, and 343 (86.6%) patients survived <20 years. A comparative study between the two groups was made. Results  By March of 2007, 67.6% (36/53) patients are still alive, disease free; 5.7% (3/53) patients died of tumor recurrence or metastasis; 11.3% (6/53) patients died of liver failure; 5.7% (5/53) patients were lost during follow-up. The longest patient survived 43 years and 2 months. Five young patients got married after resection and have had babies. One patient with a tumor measuring 17 × 13 × 9 cm (largest tumor in this series) survived for 37 years after resection, still alive, free of disease. Reresection for recurrence was done in nine patients, mean survival being 26 years and 11 months. Reresection for solitary pulmonary metastasis was carried out in three patients, mean survival being 29 years and 2 months. In comparison with patients surviving <20 years, patients surviving ≥20 years were significantly younger (P = 0.031), had a higher incidence of asymptomatic tumors (56.6 vs. 34.4%, P = 0.002); lower γ-glutamyl transpeptidase level (≤50 U/L, 64.2 vs. 25.9%, P < 0.000), lower proportion of liver cirrhosis (66.0 vs. 83.6%, P = 0.002); higher percentage of small tumors (≤5 cm, 62.3 vs. 29.9%, P < 0.000), single nodule tumors (90.6 vs. 62.9%, P < 0.000), and well-encapsulated tumors (86.8 vs. 43.6%, P < 0.000); lower proportion of tumor emboli in the portal vein (3.8 vs. 22.5%, P = 0.002), better differentiation of tumor cells (Edmondson grade I, 21.6 vs. 9.1%, P = 0.036), and higher curative resection rate (100 vs. 64.1%, P < 0.000). Conclusions  Early detection and curative resection are the principal factors improving long-term survival. Long-term follow-up after resection of HCC is very important, and should continue for the remainder of the patient’s life. Reresection for recurrence and metastasis is important approach to improve prognosis. Present in part at the Hong Kong-Shanghai International Liver Congress, 12–15 June 2008, Hong Kong.     

Use of insulin glargine and cancer incidence in Scotland: a study from the Scottish Diabetes Research Network Epidemiology Group

Aims/hypothesis  The aim of the present study was to examine whether patients with diabetes in Scotland using insulin glargine have a greater cancer risk than patients using other types of insulin. Methods  We used a nationwide diabetes clinical database that covers the majority of the Scottish population with diagnosed diabetes, and examined patients with diabetes who were exposed to any insulin therapy between 1 January 2002 and 31 December 2005. Among these we defined a fixed cohort based on exposure during a 4 month period in 2003 (n = 36,254, in whom 715 cases of cancer occurred) and a cohort of new insulin users across the period (n = 12,852 in whom 381 cancers occurred). Records from these cohorts were linked to cancer registry data up to the end of 2005. We used Cox proportional hazards models for survival analyses. Results  Those receiving any insulin glargine (n = 3,959) had the same incidence rate for all cancers as those not receiving insulin glargine (HR 1.02, 95% CI 0.77–1.36, p = 0.9 in the fixed cohort) The subset of patients using insulin glargine alone (n = 447) had a significantly higher incidence of all cancers than those using other insulins only (n = 32,295) (HR 1.55, 95% CI 1.01–2.37, p = 0.045), and those using insulin glargine with other insulins (n = 3,512) had a slightly lower incidence (HR 0.81, 95% CI 0.55–1.18, p = 0.26). There were important differences in baseline characteristics between these three groups, although the risk ratios were broadly unaltered on adjustment for these. Overall, there was no increase in breast cancer rates associated with insulin glargine use (HR 1.49, 95% CI 0.79–2.83, though insulin glargine only users had a higher rate than those using non-glargine insulin only (HR 3.39, 95% CI 1.46–7.85, p = 0.004). Among type 2 diabetic incident insulin users, no significant difference between the three groups was observed with respect to all cancer or breast cancer. All the above HRs are adjusted for age, calendar time prior cancer and type of diabetes, as appropriate, and are stratified according to sex. Conclusions/interpretation  Overall, insulin glargine use was not associated with an increased risk of all cancers or site-specific cancers in Scotland over a 4 year time frame. Given the overall data, we consider the excess of cases of all cancers and breast cancer in the subgroup of insulin glargine only users to more likely reflect allocation bias rather than an effect of insulin glargine itself. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorised users. The SDRN Epidemiology Group members involved in this study were (in alphabetical order): S. Brearley, University of Dundee; J. Chalmers, NHS Fife; H. M. Colhoun, University of Dundee & NHS Fife; S. Cunningham, University of Dundee; A. Emslie-Smith, GP & University of Dundee; C. Fischbacher, ISD Scotland; R. Lindsay, University of Glasgow; S. Livingstone, University of Dundee; R. McAlpine, University of Dundee; J. McKnight, University of Edinburgh & NHS Lothian; A. Morris, University of Dundee; D. Pearson, University of Aberdeen; J. Petrie, University of Dundee; S. Wild, University of Edinburgh An erratum to this article can be found at     

Survival analysis of patients with dermatomyositis and polymyositis

To estimate the mortality rate and identify factors predicting survival in patients with polymyositis (PM) and dermatomyositis (DM). The medical records of 192 PM/DM patients who were treated at Chang Gung Memorial Hospital from 1999 through 2008 were retrospectively reviewed. The Taiwan National Death Registry (1999–2008) was used to obtain their survival status. Thirty-one (16.1%) of the 192 patients with PM/DM had an associated malignancy; 41 (21.4%) had interstitial lung disease (ILD). During the follow-up period, 55 (28.6%) patients died and the overall cumulative survival rate was 79.3% at 1 year, 75.7% at 2 years, 69.9% at 5 years, and 66.2% at 10 years. In univariate analysis, older age at PM/DM onset, anemia, thrombocytopenia, leukopenia, diabetes mellitus, ILD, cancer, and non-use of azathioprine were associated with higher mortality (p = 0.0172, 0.0484, <0.0001, 0.0008, 0.0001, 0.0036, 0.0010, and 0.0019, respectively). In multivariate Cox regression analysis, thrombocytopenia (hazard ratio [HR] 4.94, 95% confidence interval [CI] 2.60–9.37, p < 0.0001), diabetes mellitus (HR 2.57, 95% CI 1.38–4.80, p < 0.0001), cancer (HR 2.30, 95% CI 1.26–4.22, p = 0.0030), and ILD (HR 1.98, 95% CI 1.11–3.51, p = 0.0182) were positively associated with mortality. Use of azathioprine (HR 0.35, 95% CI 0.16–0.74, p = 0.0064) was negatively associated with mortality. This study confirmed the high mortality rate (28.6%) in PM/DM patients. Survival time was significantly reduced in patients with thrombocytopenia, diabetes mellitus, ILD, and cancer patients than in those without these conditions.     

Race/Ethnicity and Risk of AIDS and Death Among HIV-infected Patients with Access to Care

BACKGROUND  Prior studies evaluating racial/ethnic differences in responses to antiretroviral therapy (ART) among HIV-infected patients have not adequately accounted for many potential confounders, and few have included Hispanic patients. OBJECTIVE  To identify racial/ethnic differences in ART adherence, and risk of AIDS and death after ART initiation for HIV patients with similar access to care. DESIGN  Retrospective cohort study. PARTICIPANTS  4,686 HIV-infected patients (66% White, 20% Black, and 14% Hispanic) initiating ART and who were enrolled in an integrated healthcare system. MEASUREMENTS  Main outcomes evaluated were ART adherence, new AIDS clinical events, and all-cause mortality. The potential confounding effects of demographics, socioeconomic status, ART parameters, HIV disease stage, and other clinical parameters were considered in multivariable models. RESULTS  Adjusted mean adherence levels were higher among White (70.1%; ref) compared with Black (64.2%; P < 0.001) and Hispanic patients (65.2%; P < 0.001). Adjusted hazard ratios (HR) for the risk of new AIDS events (White patients as reference) were 1.3 (P = 0.09) for Black and 0.9 (P = 0.64) for Hispanic patients. The adjusted HR for AIDS comparing Hispanic to Black patients was 0.7 (P = 0.11). Hispanic patients had fewer deaths compared with other racial/ethnic groups, particularly cancer and cardiovascular-related. However, adjusted HRs for death were 1.2 (P = 0.37) and 0.9 (P = 0.62) for Black and Hispanic patients, respectively, compared with White patients and 0.9 (P = 0.63) for Hispanic compared with Black patients. Adjustment for adherence did not change inferences for AIDS or death. CONCLUSIONS  In the setting of similar access to care, we did not observe a disparity for the risk of clinical events for racial/ethnic minorities, despite lower ART adherence. This material was presented at the 4th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention, Sydney, Australia, July 22–25, 2007 (#WEPEB107). This research was supported by a Community Benefit grant from Kaiser Permanente Northern California and grant number K01AI071725 from the NIAID.