Results of reoperation in 48 patients who developed recurrent brain me
tastases between January 1984 and April 1993 are presented. Median tim
e from first craniotomy to diagnosis of recurrence (time to recurrence
) was 6.7 months. Median Karnofsky performance scale (KPS) score prior
to reoperation was 80. Recurrence was local in 30 patients, distant i
n 16 patients, and both local and distant in two patients. Median surv
ival time after reoperation was 11.5 months. There were no operative m
ortalities. Multivariate analysis revealed that presence of systemic d
isease (p = 0.008), KPS scores less than or equal to 70 (p = 0.008), t
ime to recurrence of less than 4 months (p = 0.008), age greater than
or equal to 40 years (p = 0.51), and primary tumor type of breast or m
elanoma (p = 0.028) negatively affected patient survival time. These f
ive factors were used to develop a grading system (Grades I-IV). Patie
nts categorized in Grade I had a 5-year survival rate of 57%, whereas
the median survival time of patients in Grades II, LII, and IV was 13.
4, 6.8, and 3.4 months, respectively (p < 0.0001). Overall, 26 patient
s developed a second recurrence after reoperation. Seventeen patients
underwent a second reoperation, whereas nine did not. Patients undergo
ing a second reoperation survived a median of 8.6 additional months ve
rsus 2.8 months for those who did not(p < 0.0001). This study conclude
s that reoperation for recurrent brain metastasis can prolong survival
and improve quality of life. A second reoperation can also increase s
urvival, Five factors influence survival: status of systemic disease,
KPS score, time to recurrence, age, and type of primary tumor. The gra
ding system using these five factors correlates with survival time. Re
operation should be approached with caution in Grade IV patients becau
se of their poor prognosis.