When symptoms persist: choosing among alternative somatic treatments for schizophrenia.

Many patients with schizophrenia continue to have significant disabling symptoms despite adequate trials of different types and doses of traditional neuroleptics. Clinicians treating these neuroleptic-resistant patients must look to other treatments in the hope of providing some relief. The literature on many of the alternative treatments is too scanty for firm conclusions. We offer criteria for deciding which treatments may warrant consideration. We review the evidence for the eight treatments we found to meet these criteria and discuss clinical points salient to their use in this population. Although not always conclusive, the data do offer clues for treatment guidelines and an approach to choosing among the available treatments is suggested.

Anatomical abnormalities in the brains of monozygotic twins discordant for schizophrenia.

Recent neuroradiologic and neuropathological studies indicate that at least some patients with schizophrenia have slightly enlarged cerebral ventricles and subtle anatomical abnormalities in the region of the anterior hippocampus. Using magnetic resonance imaging (MRI), we studied 15 sets of monozygotic twins who were discordant for schizophrenia (age range, 25 to 44 years; 8 male and 7 female pairs). For each pair of twins, T1-weighted contiguous coronal sections (5 mm thick) were compared blindly, and quantitative measurements of brain structures were made with a computerized image-analysis system. In 12 of the 15 discordant pairs, the twin with schizophrenia was identified by visual inspection of cerebrospinal fluid spaces. In two pairs no difference could be discerned visually, and in one the twin with schizophrenia was misidentified. Quantitative analysis of sections through the level of the pes hippocampi showed the hippocampus to be smaller on the left in 14 of the 15 affected twins, as compared with their normal twins, and smaller on the right in 13 affected twins (both P less than 0.001). In the twins with schizophrenia, as compared with their normal twins, the lateral ventricles were larger on the left in 14 (P less than 0.003) and on the right in 13 (P less than 0.001). The third ventricle also was larger in 13 of the twins with schizophrenia (P less than 0.001). None of these differences were found in seven sets of monozygotic twins without schizophrenia who were studied similarly as controls. We conclude that subtle abnormalities of cerebral anatomy (namely, small anterior hippocampi and enlarged lateral and third ventricles) are consistent neuropathologic features of schizophrenia and that their cause is at least in part not genetic. Further study is required to determine whether these changes are primary or secondary to the disease.

A quantitative investigation of hippocampal pyramidal cell size, shape, and variability of orientation in schizophrenia.

Hippocampal abnormalities have been described in patients with schizophrenia, with disarray of pyramidal cells being one of the more intriguing findings. Controversy exists regarding whether disarray is present in the brains from schizophrenics in the Yakovlev collection at the Armed Forces Institute of Pathology, Washington, DC. We examined for disarray the CA1 region of the midhippocampus of 17 schizophrenics and 32 controls from this collection using computerized determination of neuronal angle and directional statistical analysis of the variability of neuronal angle. Neuronal area and shape were also assessed. We found no differences between patients and controls in these measures. Possible methodological reasons for the discrepancy between our and others' findings are discussed, as well as directions for further research into possible pathological study of the hippocampus and related structures in schizophrenia.

Predicting coronary artery disease with treadmill stress testing: changes in R-wave amplitude compared with ST segment depression.

Coronary angiograms and treadmill stress tests were reviewed independently in 108 nonconsecutively selected cases. There were 16 patients (15%) with infarcts on ECG. Changes in R-wave amplitude and ST segments during exercise were evaluated to determine the sensitivity and specificity of each as a predictor of coronary artery disease (CAD). ST segment changes had a sensitivity of 49%, and a specificity of 74%. The sensitivity increased to 55% when infarcts were excluded. R-wave amplitude changes had a sensitivity of 68% and a specificity of 84%. The sensitivity increased to 78% when infarcts were excluded. An index formed by the sum of the change in R-wave amplitude and the magnitude of ST segment change yielded a sensitivity of 76% and specificity of 78%. The sensitivity increased to 84% when infarcts were excluded. There was no statistical difference between specificities for each criteria. Of those patients with an R-wave amplitude decrease, 69% had no coronary artery atherosclerosis, while 31% had significant lesions. Of those patients with no change or an increase in R-wave amplitude, 83% had coronary artery atherosclerosis, while 17% were normal. Of the 83% with coronary artery atherosclerosis, 81% had two and three vessel disease, while only 19% had single vessel disease. No change or an increase in R-wave amplitude during treadmill stress testing is a more reliable indicator of CAD in our laboratory than ST segment changes.

Evaluation of R wave amplitude changes versus ST-segment depression in stress testing.

Exercise ECGs and coronary angiograms were reviewed in 266 patients (81 normals and 185 with significant coronary artery disease). Thirty-three false positive and 96 false negative ST responses to stress testing were purposely chosen to determine if the R wave could reduce the number of false ST responses. R wave amplitude changes were measured in the control and in the immediate postexercise period. An increase or no change in R wave was taken as evidence of an abnormal response, while a decrease in the R wave was a normal response. The sensitivity by ST segment was 48% and the specificity was 59%. These values were low because of the large number of false positive and negative ST responses in the study. It was our purpose to determine if these lowered values could be significantly improved by the R wave. Using R wave criteria, the sensitivity was 63% (P is less than 0.01) while the specificity was 79% (P is less than 0.01). The sensitivity and specificity of stress testing can be significantly improved using R wave changes.