Lawrence J. Emrich

A Method for Generating High-Dimensional Multivariate Binary Variates

Abstract Examples are given of the need for simulating correlated binary variates with different given marginal expectations and pairwise correlations. An algorithm is then presented for generating such variates. The algorithm may be used to generate variates of any dimension.

Relationship of serum antibiotic concentrations to nephrotoxicity in cancer patients receiving concurrent aminoglycoside and vancomycin therapy

Methicillin-resistant coagulase-negative staphylococci have become increasingly responsible for febrile episodes in cancer patients, often necessitating the addition of vancomycin to an aminoglycoside-containing broad-spectrum antibiotic regimen. A total of 229 courses of antibiotic therapy in 229 patients were evaluated for nephrotoxicity associated with the administration of an aminoglycoside and/or vancomycin. The incidence of nephrotoxicity observed in patients administered an aminoglycoside (Group A) was 18 percent; vancomycin (Group B) 15 percent; and an aminoglycoside concurrently with vancomycin (Group C) 15 percent. The following pharmacokinetic/dosing factors were significantly associated with increased nephrotoxicity in the groups: baseline serum creatinine level, mean daily dose during the first three days of therapy (Group B), and elevated serum trough aminoglycoside or vancomycin concentrations (2 micrograms/ml or more or 10 micrograms/ml or more, respectively). No cumulative nephrotoxicity was demonstrated with the concurrent administration of vancomycin and an aminoglycoside. A higher incidence of nephrotoxicity was seen in Group C (42 percent) and Group B (27 percent) patients, in whom trough serum vancomycin concentrations were 10 micrograms/ml or more.

Management of retroperitoneal sarcomas and patient survival

Sixty-eight patients with retroperitoneal sarcomas had an estimated 5 year survival rate of 34 percent after initial treatment. Patients with complete resection of the tumor had a 5 year rate of 64 percent and a 7 year rate of 56 percent. The respective survival rates for patients with partial excision were 33 percent and 0 and for those with only a biopsy the survival rates at 5 and 7 years were 10 percent. Radiotherapy with chemotherapy or chemotherapy alone was given to 47 patients. In the treatment of macroscopic tumor, radiotherapy or radiotherapy plus chemotherapy resulted in a 5 year survival rate of 10 percent and chemotherapy alone in a 5 year survival rate of 8 percent. Complete surgical resection is an important part of the management of retroperitoneal sarcomas and, with modern techniques of exposure, should be possible in the majority of patients.

Groin dissection in malignant melanoma

One hundred seventeen patients with malignant melanoma who had groin dissection were reviewed. The estimated 5 year survival rate for patients with node involvement was 40 percent. For patients with involved inguinal nodes only, the 5 year survival rate was 47 percent. The estimated 5 year survival rate for patients with clinically enlarged and histologically involved nodes was 37 percent and the incidence of involved deep nodes in this group was 44 percent. For patients with clinical and histologic involvement of the inguinal and deep nodes, the estimated 5 year survival rate was 30 percent. In patients with clinical involvement of the inguinal nodes, radical groin dissection with in-continuity removal of the deep nodes appeared to improve the previously reported survival rates.

Tumor thickness and prognosis in clinical stage I malignant melanoma

The current grouping of patients with malignant melanoma into thin, intermediate, and thick melanomas provides a convenient but arbitrary classification which, although providing “average” survival values for each group, offers crude prognostication for the individual patient. A review of 371 patients with clinical Stage I malignant melanoma, treated during the period 1970 to 1985, was conducted. The estimated 5‐year survival rate for female patients with melanomas 1.0 mm thick was 94%; for each 1‐mm increment in thickness the survival rate declined by about 3%, up to the 6 mm mark, the survival rate declining thereafter by about 8% for each additional millimeter in the range of 7 to 15 mm of thickness. The estimated 5‐year survival rate for male patients with melanomas 1.0 mm thick was 80%; for each 1‐mm increment the survival rate declined by about 9%, up to the 10 mm mark. The proposed method of estimating the expected survival according to the patients sex and the thickness of the primary lesion hopefully provides a more accurate and convenient method of prognostication for the clinician dealing with specific patients with intermediate or thick melanomas.

Variants of hemipelvectomy and their complications

In the period from 1976 through 1986, 62 procedures were performed. Of these, 42 were posterior flap hemipelvectomies, 5 anterior flap hemipelvectomies, and 15 internal hemipelvectomies. The median duration of these procedures was 6.5 hours, and the median blood loss was 2,541 ml. Postoperatively, there were no wound problems in 38 procedures (61 percent). The overall rate of flap necrosis was 15 percent, and the overall rate of wound infection, 17 percent. The viability of the posterior flap was not dependent on the level of division of the iliac vessels. By leaving the gluteus maximus muscle attached to the posterior flap, the rate of flap necrosis, initially 55 percent, was eliminated completely in the last 38 patients. Including 11 recently performed procedures, the operative mortality rate was 1 percent (1 of 73 procedures). For patients operated on with curative intent, the estimated 5-year survival rate was 43 percent.

Effect of surgical treatment on stage IV melanoma

One hundred eighty patients with hematogenous metastases from malignant melanoma were reviewed. Complete resection of the gross tumor was technically feasible in 33 percent of the cases. Patients who had complete resection of the gross tumor had an estimated median survival time of 11.4 months and an estimated 5 year survival rate of 14 percent. Patients with solitary lesions removed had a median survival time of 22.8 months and a 5 year survival rate of 23 percent. Patients with distant subcutaneous metastases completely removed had a median survival time of 31.9 months and a 5 year survival rate of 29 percent. Surgical resection of distant metastases, when technically feasible, particularly for solitary lesions and subcutaneous locations, increases the length of survival of patients with disseminated melanoma.

Hydroxyurea plus pelvic radiation versus placebo plus pelvic radiation in surgically staged stage IIIB cervical cancer

Forty‐five evaluable patients with stage IIIB carcinoma of the uterine cervix were entered into a prospective, double‐biind, randomized study to evaluate the possible radiation‐potentiating properties of hydroxyurea. All patients were documented to be without para‐aortic lymph node metastasis by pretherapy staging para‐aortic lymphadenectomy. The original plan of therapy was for continuous therapy (200 rads/day) of 6,000 rads of pelvic radiation for 6 weeks plus intrauterine radium. However, 16 patients received 6,000 rads in 8 weeks by split‐course therapy (2‐week rest after 3,000 rads) plus radium. Twenty‐nine patients received the planned continuous therapy. The median dose of pelvic radiation for patients who received continuous therapy or split‐course radiation was 6,000 rads. Leukopenia (WBC < 2,500/mm3) was significantly increased in the patients given hydroxyurea as compared to those given placebo (P < .001). There was no statistically significant difference relative to anemia, thrombocytopenia, radiation skin reaction, diarrhea, or radiation‐induced complications requiring surgical correction. The estimated 5‐year progression‐free survival rate for the combined, continuous, and split‐course radiation therapy hydroxyurea patients was 60 %, and its was 52 % for the corresponding placebo patients (P = .49). However, the estimated 5‐year progression free survival rate for the correctly treated patients (continuous therapy) was 91% for the hydroxyurea group and 60% for the placebo group (P < .06).

Prognostic significance of lymph node metastasis and bone, major vessel, or nerve involvement in adults with high-grade soft tissue sarcomas

Two hundred sixty‐seven patients with high‐grade (G2 or G3) soft tissue sarcomas (STS) were examined. All tumors were removed by resection (marginal or wide) or amputation. Seventy‐four patients had T3 primary tumors invading neurovascular structures (n = 41) or bone (n = 33), and 29 patients had histologically confirmed metastases to the regional nodes removed at the time of definitive surgery. The estimated 5‐year and 10‐year survival rates for patients in Stage IIa, b or IIIa, b were 44% and 37%, respectively. For patients with neurovascular or bone invasion the survival rates were 24% and 15%, respectively. For patients with lymph node metastases the survival rates were 10% and 3%, respectively. Survival of patients with primary sarcomas invading the nerve, vessel, or bone was significantly better than that of patients with lymph node metastases (P = 0.002). Survival also was distinctly different between patients with nerve or vessel invasion who had a 5‐year survival rate of 32%, and patients with bone invasion who had a 5‐year survival rate of 15% (P = 0.002). These findings suggest that the current staging system for STS should be reexamined. Also, patients with nerve or vessel invasion should be assigned a IIIc, position, those with bone invasion a IIIc2 position, and those with lymph node metastases a IVa position in the staging system.