Pl. Penar et Jt. Wilson, COST AND SURVIVAL ANALYSIS OF METASTATIC CEREBRAL-TUMORS TREATED BY RESECTION AND RADIATION, Neurosurgery, 34(5), 1994, pp. 888-893
THE SURGICAL TREATMENT of metastatic brain tumors remains controversia
l, primarily because of the limited prognosis of patients with metasta
tic cancer. The cost effectiveness of even standard therapies is of in
creasing concern to third-party payers. We reviewed the records of pat
ients who had a single metastatic brain tumor resected at the Medical
Center Hospital of Vermont (a referral center in a rural state) since
cost data recording began. The 32 patients ranged in age from 35 to 77
years, with a 2.2:1 female-to-male ratio. Thirty-four percent of tumo
rs originated in the lung, 15.6% were renal, 12.5% were breast, 12.5%
were gynecological, 9.4% were gastrointestinal, and 9.4% were ultimate
ly of unknown origin. Thirty-one tumors were completely resected; 30 p
atients were irradiated, most after surgery (mean dose, 3,908 +/- 1,25
0 cGy). Karnofsky scores improved from 80 +/- 11 to 88 +/- 16 postoper
atively (P = 0.0038, one-tailed paired t-test). Patients were hospital
ized an average of 8.22 +/- 6.26 days postoperatively, with total oper
ative and postoperative charges of $19,190 +/- 5,684, noninclusive of
radiotherapy. The expected median survival of all patients was 26 mont
hs (Kaplan-Meier estimate). The presence of disseminated disease was n
ot significantly correlated with survival (P = 0.237). The number of p
ostoperative days was more for patients with disseminated disease (P =
0.0274), but not for patients with infratentorial tumors (P = 0.6991)
. Age higher than the median was not correlated with an increased numb
er of postoperative days (P = 0.1366) nor was a preoperative Karnofsky
score of 70 or less (P = 0.1382). We believe that the cost of adding
surgical therapy is not out of proportion to the degree of palliation
achieved, especially because surgical treatment can result in rapidly
improved function and long-term survivals are not uncommon. Factors su
spected to influence the length of hospitalization such as age and tum
or location do not affect postoperative stay and thus the cost of trea
tment.