Unfortunately, I haven't found able to find either my old notes on research on the treatment of psychotic depression or very many of the double-blind studies on PubMed looking at the benefit, if any, of adding an anti-psychotic to an anti-depressant in the treatment of psychotically depressed individuals.  I do, however, recall that there are several older well-done studies in addition to those below showing that for most patients adding an anti-psychotic is of no benefit in treating psychotic depression.  Modern studies continue to confirm the findings of older studies.

Olanzapine plus Fluoxetine (Symbyax) Not a Worthwhile Treatment for Psychotic Depression: In two 8-week DB PC studies, 249 patients with Major Depressive Disorder with psychotic elements were given olanzepine (5 to 20 mg/d), placebo, or olanzapine 5 to 20 mg/d + fluoxetine 20 to 80 mg/d. In the first study, olanzapine did not do significantly better than the placebo (-14.9 vs. -10.4 Hamilton Depression score), but the combination did (-20.9 vs. -10.4)( P = 0.001); The OFC group also had significantly higher response rate (63.6%) than the placebo (28.0%) or olanzapine (34.9%) groups. In the second trial, no group did significantly better than another. A Double-Blind, Randomized Study of Olanzapine and Olanzapine/Fluoxetine Combination for Major Depression With Psychotic Features. Rothschild AJ, Williamson DJ, Tohen MF, Schatzberg A, Andersen SW, Van Campen LE, Sanger TM, Tollefson GD. University of Massachusetts, Eli Lilly Drug Company, Harvard, and Stanford Universities. J Clin Psychopharmacol. 2004 Aug;24(4):365-373. Ed: It is sad to see Harvard and Stanford colluding with Eli Lilly in this intentionally poorly designed study.  The Lilly company was very careful to stack the deck in favor of its new combination medication Symbyax by not including a fluoxetine only group in the study.  It is very unlikely that olanzapine is of any benefit in treating the large majority of psychotically depressed patients.  Of course, plain fluoxetine costs only $2 per month vs. $225 per month for Lilly's new Symbyax. Symbyax causes considerable weight gain and a fair number of cases of diabetes, while fluoxetine alone does not.

Psychotic Depression: Start with Anti-Depressant Alone: In a meta-analysis of 10 randomised controlled trials, researchers found no evidence that the combination of an antidepressant with an antipsychotic is more effective than an antidepressant alone. Antidepressant monotherapy and adding an antipsychotic if the patient does not respond, or starting with the combination of an antidepressant and an antipsychotic, both appear to be appropriate options for patients with unipolar psychotic depression. However, clinically the balance between risks and benefits may suggest the first option should be preferred for many patients. Starting with an antipsychotic alone appears to be inadequate. Pharmacological treatment for unipolar psychotic depression: Systematic review and meta-analysis. Wijkstra J, et al. University Medical Centre Utrecht, The Netherlands. . Br J Psychiatry 2006 May;188:410-5.

Antipsychotic No Added Benefit to Nortriptyline in Late-Life Psychotic Depression: In a DB PC trial of 36 psychotically depressed adults over age 49, all received nortriptyline and half received perphenazine with the others getting placebos for an average of 9 weeks. There was no significant difference between the completers in the 2 treatment groups when comparing their scores on the HAM-D, the BPRS, its psychoticism subscale, or any side effects measure. Rates of response were nortriptyline-plus-perphenazine 50% vs. nortriptyline-plus-placebo 44%, not a significant difference. A double-blind randomized comparison of nortriptyline plus perphenazine versus nortriptyline plus placebo in the treatment of psychotic depression in late life. Mulsant BH, Sweet RA, Rosen J, Pollock BG, Zubenko GS, Flynn T, Begley AE, Mazumdar S, Reynolds CF 3rd. University of Pittsburgh. J Clin Psychiatry. 2001 Aug;62(8):597-604

ECT Relapse: Nortriptyline Better than Nortriptyline + Perphenazine: 28 psychotic depressed had ECT. Those on nortriptyline alone to prevent relapse did non-significantly better and had fewer side-effects. Continuation treatment of delusional depression in older adults. Meyers BS, Klimstra SA, Gabriele M, Hamilton M, Kakuma T, Tirumalasetti F, Alexopoulos GS. Am J Geriatr Psychiatry 2001 Fall;9(4):415-22

Mifepristone Successful in Psychotic Depression: Five patient DB PC crossover 4 days 600 mg. 4 of 5 substantially improved. Higher dose might be possible. Rapid reversal of psychotic depression using mifepristone. Belanoff JK, Flores BH, Kalezhan M, Sund B, Schatzberg AF. Department of Psychiatry, Stanford. J Clin Psychopharmacol 2001 Oct;21(5):516-21

Imipramine Alone Fine with Mood Congruent Depressive Psychosis: In a DB PC study of 52 patients of whom one-quarter were psychotic, the psychotically depressed patients had a 70% response and actually did better than non-psychotically depressed. Treatment of mood-congruent psychotic depression with imipramine. Bruijn JA, Moleman P, Mulder PG, van den Broek WW. Department of Psychiatry, University Hospital Rotterdam. J Affect Disord 2001 Oct;66(2-3):165-74

Psychotic Depression Responds to Anti-Depressants Alone: In a study of 150 in-patients with Major Depressive Disorder, patients were treated with imipramine or amitriptyline up to 250 mg/day for 4 weeks. Severe non-psychotic and psychotic depression responded less well (39% and 32% had good responses) than those with moderate depression (67% responded well) but equally well to each other in 4 week study. Response to treatment with antidepressants of patients with severe or moderate non-psychotic depression and of patients with psychotic depression. Kocsis JH, Croughan JL, Katz MM, Butler TP, Secunda S, Bowden CL, Davis JM. Am J Psychiatry 1990 May;147(5):621-4