Predictors of Depression and Anxiety in Patients with Intracranial Neoplasms

D epression rates in patients with intracranial neoplasms have been reported to range from 19% to 93%, with most studies reporting prevalence rates between 10% and 30%. 1 – 4 Several studies have examined characteristics of depressed patients with brain tumors, 1 , 3 – 5 but few have examined the presence and characteristics of anxiety in this population. 3 , 6 The comorbidity of having a psychiatric illness with an intracranial tumor has been shown to result in multiple complications. 7 , 8 Although there is an extensive literature examining the relationships between depression, anxiety, and cancer, few systematic studies have addressed their occurrence specifically in association with brain tumors.

METHOD

A consecutive series of deceased patients with diagnoses of primary or metastatic brain cancer who were admitted to the Johns Hopkins Hospital between September 1995 and January 2007 was obtained using billing codes. Fifty-six cases were excluded due to inadequate records or because the patient was under 5 years of age. Medical records were examined for demographic data, tumor type, location, size, treatment modalities, survival time, and premorbid psychiatric diagnoses. Treating physicians’ diagnoses of depression and anxiety or data in the chart that were suggestive of either disorder according to DSM-IV criteria, were noted. Diagnostic or severity scales were not utilized in the assessment of affective disorders. The use of antidepressants and concurrent treatment by a psychiatrist were also recorded. Data were analyzed using chi-square and ANOVA analyses using SPSS 15.0 software (Chicago, Ill.). This study was approved by the Johns Hopkins Hospital Institutional Review Board.

RESULTS

Table 1 summarizes characteristics of the study population (N=301). There was an even cerebral distribution of single site tumors. Ninety-four (31%) individuals had bilateral tumors. Depression was the most common premorbid psychiatric disorder (n=18, 6%) followed by substance abuse (n=17, 6%) and anxiety disorders (n=16, 5%). Comorbid depression was diagnosed in 45 individuals (15%), 35 individuals had anxiety disorders (12%), and 13 patients had both disorders following the diagnosis of a brain tumor (4%). Thirty-four depressed patients (75%) were prescribed antidepressants, and 12 were seen by a psychiatrist at some point during their illness (27%). Depression was reported as frequently in males (n=21, 47%) as in females (n=24, 53%), but diagnoses of anxiety were more prevalent in females (n=26, 74%, p=0.01). Race, marital status, tumor type, location, and size were not found to be predictive of either disorder. Premorbid depression (n=18, 40%, p<0.001) and anxiety (n=16, 46%, p<0.001) were found to be highly predictive of future psychiatric illness in the setting of brain tumors using bivariate analyses. Survival times varied for patients with depression (11 months), anxiety (7 months), both disorders (3.5 months), and neither disorder (9 months), but were not statistically significant (p=0.69). Thirty patients with depression (67%, p=0.067) were given whole brain radiation treatments while 22 individuals with anxiety (63%) received radiation therapy (p=0.058). Eight patients with both disorders (62%) underwent whole brain radiation treatment.

Enlarge table

DISCUSSION

Depression and anxiety rates, as detected by internists and oncologists in this study, were similar to the 12-month prevalence rate in the general population. 9 The rate of comorbid anxiety and depression (4%) was less than expected. However, many studies have found underestimation of depression and anxiety in the medically ill. 10 This and the lack of a formal concurrent psychiatric assessment suggest that the prevalence rates for these disorders may not be accurately reported. Interestingly, reported depression rates were relatively equivalent in both genders, a finding that is contradictory to the 2:1 ratio of female to male depression rates noted in both general and stroke population studies. 9 , 11 This could be attributed to the erroneous notion that the equal prevalence rates of brain tumors across gender relate to equal prevalence rates of depressive episodes in both males and females. In the general population, the female to male ratio for anxiety disorders is also approximately 2:1 9 ; however, the ratio neared 3:1 in this study. This suggests that females with brain tumors may be particularly susceptible to anxiety disorders, though this may also be attributable to gender bias. 12

In this study, the most common correlate of reported depression and anxiety in brain tumor patients was a premorbid diagnosis of either disorder. Sixty-one percent of patients with a premorbid diagnosis of depression were deemed depressed after being diagnosed with an intracranial neoplasm while 63% of premorbidly anxious patients had a recurrence of symptoms, both of which were statistically significant findings. Because the current study examined physician-reported rates of psychiatric illness, it is possible that clinicians were more apt to diagnose patients who had previously been diagnosed with a psychiatric illness. Nonetheless, we recommend that a past psychiatric history be obtained when evaluating patients with brain tumors and that those patients with a history of depression or anxiety be monitored closely. Further studies should address whether prophylactic antidepressant therapy is warranted as this has been shown to significantly reduce depression rates in patients following stroke. 13 Although the results did not reach statistical significance in this study, data suggest that survival time may be shortened by depression and anxiety as previously reported. 2 , 7

There are several limitations to this study. First, data were derived from records that may not have accurately portrayed patient symptoms, as standardized psychiatric assessments were not performed. This may have resulted in erroneous documentation of comorbid and premorbid psychiatric diagnoses. Because the study was conducted at a tertiary care hospital, this may also have biased tumor type and illness severity.

In summary, psychiatric illnesses in brain cancer patients may be underdiagnosed and undertreated for a variety of reasons, 14 , 15 possibly resulting in decreased survival times in patients with anxiety and in those with both depression and anxiety. Routine use of screening tools for depression and anxiety may improve the detection rate of these two illnesses within the neurooncological setting. Premorbid depression and anxiety are the strongest predictors of subsequent psychiatric illness and females may be especially susceptible to anxiety symptoms. Patients exposed to whole brain radiation treatment may also be at increased risk of developing a psychiatric illness, as has been previously suggested. 16