Redalyc.Oral Fluphenazine Versus Placebo for Schizophrenia

Revista Colombiana de Psiquiatría ISSN: 0034-7450 [email protected] Asociación Colombiana de Psiquiatría Colombia

El-Din Matar, Hosam; Qutayba Almerie, Muhammad; Giraldo, Ana María; Adams, Clive E. Oral fluphenazine versus placebo for schizophrenia: a Cochrane systematic review of 40 years of randomised controlled trials Revista Colombiana de Psiquiatría, vol. XXXVI, núm. 1, 2007, pp. 8-17 Asociación Colombiana de Psiquiatría Bogotá, D.C., Colombia

Available in: http://www.redalyc.org/articulo.oa?id=80636102

How to cite Complete issue Scientific Information System More information about this article Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal Journal's homepage in redalyc.org Non-profit academic project, developed under the open access initiative Artículoso r i g i n a l e s Oral fluphenazine versus placebo for schizophrenia: a Cochrane systematic review of 40 years of randomised controlled trials

Hosam El-Din Matar1 Muhammad Qutayba Almerie1 Ana María Giraldo2 Clive E. Adams3

Abstract:

Background: Fluphenazine, one of only three antipsychotics on WHO’s list of essential drugs, has been widely available for five decades. Quantitative reviews of its effects compared with placebo are rare and out of date. Methods: We searched for all relevant randomised controlled trials comparing oral administration of fluphenazine with placebo on the Cochrane Schizo- phrenia Group’s register of trials (October 2006) and in reference lists of included studies. Data were extracted from reliably selected trials. Where possible, we calculated fixed effects relative risk (RR), the number needed to treat (NNT), and their 95% confidence intervals (CI). Results: We found over 1200 electronic records for 415 studies. Ninety papers were acqui- red; 59 were excluded and the remainder were reports of the seven trials we could include (total participants=349). Compared with placebo, in the short-term, global state outcomes for ‘not improved’ were not significantly different (n=75, 2 RCTs, RR 0.71 CI 0.5 to 1.1). There is evidence that oral fluphenazine, in the short term, increases a person’s chances of experiencing extrapyramidal effects such as akathisia (n=227, 2 RCTs, RR 3.43 CI 1.2 to 9.6, NNH 13 CI 4 to 128) and rigidity (n=227, 2 RCTs, RR 3.54 CI 1.8 to 7.1, NNH 6 CI 3 to 17). We found study attrition to be lower in the oral fluphenazine group, but data were not statistically significant (n=227, 2 RCTs, RR 0.70 CI 0.4 to 1.1). Conclusions: Fluphenazine is an imperfect treatment with surprisingly few data from trials to support its use. If accessible, other inexpensive drugs, less associated with adverse effects, may be a better choice for people with schizophrenia. It is time for the World Health Organisation to revise their list of essential antipsychotic drugs.

Key words: Fluphenazine, schizophrenia, antipsychotics.

Título: Flufenazina oral vs. placebo, en el tratamiento de la esquizofrenia. Una revisión sistemática de Cochrane de 40 años de estudios controlados aleatorizados.

1 Medical Student, Faculty of Medicine, Damascus University, Syria. 2 Medical Student, Pontificia Universidad Javeriana, Bogotá. 3 Cochrane Schizophrenia Group, Academic Unit of Psychiatry and Behavioural Sciences, University of Leeds, UK.

8 Revista Colombiana de Psiquiatría, vol. XXXVI / No. 1 / 2007 Oral fluphenazine versus placebo for schizophrenia: a Cochrane systematic review of 40 years...

Resumen people with schizophrenia (1). They are generally regarded as highly Introducción: La flufenazina, uno de los tres antipsicóticos de la lista de drogas esencia- effective, especially in controlling les de la OMS, está disponible desde hace 5 symptoms such as hallucinations décadas. Las revisiones cuantitativas de sus and fixed false beliefs (delusions) efectos vs. placebo son escasas y desactuali- (2). Fluphenazine, a phenothiazine zadas. Método: Buscamos estudios controlados aleatorizados relevantes que compararan derivative, is one of the first drugs la administración de flufenazina oral contra to be classed as an ‘antipsychotic’. placebo, en los registros de ensayos de los Reports from 1959 and 1960 first grupos de esquizofrenia de Cochrane (oct. indicate its value in psychotic ill- 2006) y en referencias incluidas en los estu- ness (3-4) and it was approved for dios. Los estudios consistentes y confiables muestran datos de ensayos clínicos aleatori- clinical use in the USA in 1959. zados (ECA). Calculamos el RR mediante un Fluphenazine is thought to eli- modelo de efectos fijos, el número necesario cit its antipsychotic effects via inter- por tratar (NNT) y su intervalo de confianza ference with central dopaminergic de 95%. Resultados: En flufenazina vs. placebo, a corto plazo, el resultado “no mejoría” pathways and blocking receptors, no fue muy diferente (n=75, 2 ECA, RR 0,71, particularly D2, in the mesolimbic IC 0,5 a 1,1). La flufenazina oral, a corto zone of the brain (5). Extrapyra- plazo, aumenta el riesgo de manifestaciones midal side effects are a result of extrapiramidales, como la acatisia (n=227, interaction with dopaminergic 2 ECA, RR 3,43, IC 1,2 a 9,6, NNT 13, IC 4 a 128) y rigidez (n=227, 2 ECA, RR 3,54, IC pathways in the basal ganglia. As 1,8 a 7,1, NNT 6, IC 3 a 17). La deserción fue fluphenazine is not specific to one más baja en el grupo de la flufenazina oral, action within the body it is known pero los datos no fueron estadísticamente to cause adverse effects ranging significativos (n=227, 2 ECA, RR 0,70 CI 0,4 a 1,1). Conclusión: La flufenazina es un from orthostatic hypotension as a tratamiento imperfecto con muy pocos datos result of its alpha adrenergic bloc- que sustenten su uso. Otras drogas de bajo king activity to anticholinergic and costo y pocos efectos adversos pueden ser extrapyramidal symptoms (tardive mejor opción para pacientes psicóticos. La dyskinesia, pseudo-parkinsonism, OMS debe revisar su lista de antipsicóticos esenciales. dystonia, dyskinesia, akathesia) (6). In addition, the use of fluphenazine Palabras clave: flufenazina, esquizofrenia, has been associated with a poten- antipsicóticos. tially fatal disturbance of blood pressure, temperature and muscle Background control (neuroleptic malignant syndrome)(7) . Overall, the antipsychotic drugs, Fluphenazine is still commonly with their anti-dopaminergic effects, used for people with schizophre- are the mainstay treatment for nia and is given by mouth or by

Revista Colombiana de Psiquiatría, vol. XXXVI / No. 1 / 2007 9 El-Din Matar H., Almerie MQ., Giraldo A., Adams C. short-acting injection. Although ridin* or mirenil*) in title, abstract we have not found precise data on and index fields in REFERENCE) or world wide use, fluphenazine, along (fluphenazin* in interventions field with only two other antipsychotics in STUDY)) (chlorpromazine and haloperidol) is This register is compiled by re- on the World Health Organization’s gular, systematic, searches of major list of essential drugs (8). In low and bibliographic databases, hand sear- middle income countries, where ches and conference proceedings. non-proprietary preparations of flu- A full description is given in the phenazine are inexpensive, it may Group’s module on the Cochrane be one of the only drug treatments Library (http://www.cochrane.org/ available. We, however, know of no contact/entities.htm#CRGLIST). In up-to-date systematic reviews of the addition, the references of all identi- absolute effects of this ‘essential’ fied studies were inspected for more antipsychotic. studies.

Methods Data extraction and study appraisal All electronic records identified Inclusion criteria were independently inspected by The inclusion criteria were de- HEM and MQM, who then obtained fined and disseminated for peer re- full reports of studies of agreed rele- view within a Cochrane protocol (9). vance. We assessed the methodolo- We included articles if they reported gical quality of included trials using randomized controlled trials where criteria described in the Cochrane participants had schizophrenia or Handbook (10). These criteria are non-affective serious/chronic men- based on the evidence of a strong tal illness, and where the interven- relationship between allocation con- tions included oral administration cealment and direction of effect (11). of fluphenazine (any dose) versus Data relating to methods, partici- placebo or no treatment. pants, interventions and outcomes, were extracted and disagreement Identification of relevant trials discussed and documented. We identified relevant randomised trials by searching the Cochrane Statistical methods Schizophrenia Group’s Register of trials (Oct 2006) using the phrase: Dichotomous and continuous (((fluphen* or flufen* or modec* data were not used if over half of or moditen* or eutimax* or prolixin* those randomised were not con- or siqualon* or anaten* or dapotum* tributing to the outcome due to or decazate* or decafen* or decen- early attrition from the study or tan* or fludecate* or lyogen* or lyo- non-compliance. Dichotomous data

10 Revista Colombiana de Psiquiatría, vol. XXXVI / No. 1 / 2007 Oral fluphenazine versus placebo for schizophrenia: a Cochrane systematic review of 40 years... were combined using fixed effects way as to make the outcomes unin- Relative Risk (RR) (12). Numbers telligible or impossible to use. needed to treat (NNT) (13) were also calculated and I-square tests Study quality for heterogeneity were performed All seven included studies either (14). Where less than 50% of people reported use of random allocation or were lost to follow-up at the end of a suggested it; one study reported the trial, ‘worst case’ intention-to-treat process used as the groups were analyses were undertaken by assu- matched on age, chronicity and ming that those who had left a trial severity of illness (16). Citations to early had had a poor outcome. The all included and excluded studies sensitivity of the final results to this are available in the full Cochrane assumption was tested. Continuous Review (10), otherwise the names data were excluded if derived from and dates cited in this text relate scales of unknown validity and if to Table 1. Only two of the seven totals or measures of variance were included studies (29%) adequately not reported. Summation was not describe attempts to double blind attempted if continuous data were (17-18). Other trials indicated that too skewed (15). All estimates of blinding had been made, but gave effect are presented with their 95% no description of how this had confidence intervals (CI). been done. Often the description of participants who left studies early Results was inadequate; two of the seven included studies provided details Search results of treatment withdrawals. Three Electronic searches identified studies reported that withdrawal over 1200 records most of which from treatment had occurred, but were ineligible. Full copies of 90 gave no further description. Presen- possibly relevant citations were ob- tation of data was also poor. Trials tained for detailed scrutiny. Of the- frequently presented both dichoto- se, fifty-nine papers were excluded mous and continuous data in gra- and thirty-one reports of the seven phs, or reported inexact statistical randomized trials included (see Ta- measures of probability, for example ble 1). Studies were mainly excluded p>0.05. This often made it impos- due to lack of random allocation. sible to extract raw data for syn- Four randomised trials, however, thesis. Continuous scale data were reported irrelevant outcomes, such frequently collected in the trials but as critical flicker fusion frequency, were frequently poorly reported: two the effects of giving ‘phenothiazines’ of the seven included trials did not not broken down by each treatment report standard deviations and four group, or presented data in such a other included trials did not present

Revista Colombiana de Psiquiatría, vol. XXXVI / No. 1 / 2007 11 El-Din Matar H., Almerie MQ., Giraldo A., Adams C.

Tabla 1. Included studies

Studies Methods Participants Interventions Outcome Sex Death Relapse Historia (mg/day) Age (years) use placebo Randomised Double-Blind Early desertion Duration (weeks) Only scuizophrenia Total of participants Dose of fluphenazine Dose of fluphenazine General improvement received fluphenazine Number of persons who Numbers of persons who

1999 Carpenter ● ● 6 ● C 53 M=37 M+F 15 18 20 ● 1971 Clark ● ● 6 ● C 76 M=33 M+F 2-10 18 19 ● ● 1964 Goldberg ● ● 6 ● A 463 16-54 M+F 1-16 92 98 ● 1964 Hordern U/K ● 12 ● C 75 M=49 F <14 25 25 ● ● 1994 Marder ● ● 104 ● A+C 36 M=40 U/K 10 17 19 ● ● ● 1963 Millar ● ● 6 ● C 38 28-58 F 2.5 19 19 ● 1976 Rifkin ● ● 52 C 73 17-40 M+F 5-20 28 22 ● ● ●

● = Yes A = Acutely ill C = Chronically ill M = Male F = Female

any data from the scales they had community. Five studies were con- used. In this way a lot of potentially ducted in the USA, one in Australia informative data were lost. and one in the United Kingdom. All trials compared oral fluphenazine Study designs with inactive placebo. The lowest These studies included 814 par- dose of fluphenazine tested was 2.5 ticipants but only 349 of whom were mg/day (18) while the highest was allocated to the specific comparison 15 mg/day (19). The mean duration of interest to this review (oral flu- of treatment was about 170 days phenazine vs placebo). The great (~6 months), but this was highly majority of participants in nearly all skewed (SD 253). The most common trials were diagnosed as suffering study length was six weeks but the from schizophrenia. Four of the range was considerable with the seven trials described the diagnos- longest being 2 years. tic criteria used, or the symptoms required for people to be included. Outcomes Otherwise entry to most of the in- Table 2 presents the main re- cluded studies was based on a prag- sults of this review. These inten- matic diagnosis of schizophrenia. tion-to-treat data are derived by The trials ranged in size from 36 to synthesising homogeneous trial fin- 190 participants. Most people were dings and results remain essentially hospitalised at the time of the study. unchanged when we only use data Four studies were hospital-based, from participants who completed while three were undertaken in the studies. These data show no clear

12 Revista Colombiana de Psiquiatría, vol. XXXVI / No. 1 / 2007 Oral fluphenazine versus placebo for schizophrenia: a Cochrane systematic review of 40 years...

Table 2: Results relating to clinical change and study attrition

Months Number Fluphenazine Placebo RR (95% CI) I2 test for of trials n/N n/N heterogeneity No global improvement 2-6 3 34/61 40/64 0.89 (0.67, 1.18) 26.5%

Relapse 6-24 3 17/64 38/60 0.44 (0.30, 0.67) 88.0%

Leaving the study early 6-24 5 30/180 42/183 0.75 (0.50, 1.13) 0% pattern indicative of publication RCTs, RR 0.89 CI 0.67 to 1.18) with bias when sorted by study size and low heterogeneity (I2 26.5%, Figure effect (20). None of the included 1). One study reported data on re- studies attempted to quantify levels lapse up to six weeks (short term as- of satisfaction or quality of life and sessment) with results indicating a there is no evidence of any direct trend favouring fluphenazine (n=38, economic evaluation of fluphenazi- RR 0.25 CI 0.1 to 1.0). Two other ne. We are, however, able to report studies reported data for long-term some data on the absolute effects relapse which significantly favoured of oral fluphenazine on aspects of fluphenazine but data are heteroge- the global and mental state, and neous (I-squared 92%). adverse effects. Fluphenazine has many adverse Data on global improvement effects (see Table 3). Extrapyrami- (not improved or worsened), in the dal symptoms are common. In the period up to 6 months showed no short term, there is evidence that significant difference between oral fluphenazine increases a person’s fluphenazine and placebo (n=125, 3 chances of experiencing akathisia

Figure 1. Oral ﬂuphenazine vs placebo – global outcome: Not improved or worsened

Sub-Category Fluphenazine Placebo RR (ﬁxed) Weight RR (ﬁxed) Study n/N n/N 95% Cl % 95% Cl

Short term - measured by CGI Clark 1971 5/18 9/19 22.44 0.59 (0.24, 1.42) Carpenter 1999 10/18 14/20 33.99 0.79 (0.48, 1.31) Subtotal (95% Cl) 36 39 56.43 0.71 (0.45 - 1.12) Total events: 15 (Fluphenazine), 23 (placebo) Test for heterogeneity: Chi2 = 0.37, df = 1 (P = 0.55), l2 = 0% Test of overall effect: Z = 1.48 (P = 0.14)

Medium term - measured by MDRS Hordern 1964 19/25 17/25 43.57 1.12 (0.79, 1.58) Subtotal (95% Cl) 25 25 43.57 1.12 (0.79, 1.50) Total events: 19 (Fluphenazine), 17 (placebo) Test for heterogeneity: not applicable Test of overall effect: Z = 0.63 (P = 0.53)

TOTAL 61 64 100.00 0.89 (0.67, 1.18) Total events: 34 (Fluphenazine), 40 (placebo) Test for heterogeneity: Chi2 = 2.72, df = 2 (P = 0.26), l2 = 26.5% Test of overall effect: Z = 0.82 (P = 0.41)

0.2 0.5 1 2 5 Favours ﬂuphenazine Favours placebo

Revista Colombiana de Psiquiatría, vol. XXXVI / No. 1 / 2007 13 El-Din Matar H., Almerie MQ., Giraldo A., Adams C.

Table 3: Adverse effects

Number Fluphenazine Placebo RR (95% CI) of trials n/N n/N

General symptoms

Cardiovascular symptoms 1 11/92 5/98 2.34 (0.85, 6.49) Gastrointestinal distress + nausea 2 5/110 6/117 0.90 (0.30, 2.72) Lactation 1 3/92 0/98 7.45 (0.39, 42.32) Rash 2 2/110 3/117 0.76 (0.15, 3.78)

Extrapyramidal symptoms

Akathisia 2 14/110 4/117 3.43 (1.23, 9.56) Dystonia 1 24/25 22/25 1.09 (0.92, 1.29) Insomnia 2 37/110 40/117 0.99 (0.69, 1.40) Parkinsonism 1 11/25 2/25 5.5 (1.36, 22.32) Rigidity 2 30/110 9/117 3.54 (1.76, 7.14) Tremor 2 16/110 5/117 3.19 (1.25, 8.11)

(n=227, 2 RCTs, RR 3.43 CI 1.2 (weakness, faintness, dizziness), to 9.6, NNH 13 CI 4 to 128), facial lactation and rash. rigidity (n=190, RR 2.77 CI 1.0 to In the short term people allo- 7.5, NNH 12 CI 4 to 654), ‘loss of cated to oral fluphenazine did not associated movements’ (n=190, RR leave the study any less often than 6.39 CI 2.0 to 21.0, NNH 7 CI 2 to participants who were given placebo 35), rigidity (n=227, 2 RCTs, RR (n=227, 2 RCTs, RR 0.70 CI 0.4 to 3.54 CI 1.8 to 7.1, NNH 6 CI 3 to 1.1). This also applied to the me- 17) and tremor (n=227, 2 RCTs, RR dium term (n=50, 1 RCT, RR 5.0. CI 3.19 CI 1.3 to 8.1, NNH 11 CI 4 to 0.3 to 99.2) and long term follow-up 94). We found measures of akinesia, (n=86, 2 RCTs, RR 0.69 CI 0.2 to associated movements, dystonia 2.0). Overall, across all time periods, and restlessness/insomnia were only about 15% of people left these not significantly different from those studies early (n=363, 5 RCTs, RR allocated to placebo. Evidence in the 0.75 CI 0.5 to 1.1). Only one study medium term indicates that fluphe- Rifkin (21-24) reported the outcome nazine increases the likelihood of of death, with one occurring in the having parkinsonism (15) (n=50, RR fluphenazine group during long- 5.50 CI 1.4 to 22.3, NNH 3 CI 2 to term follow-up (n=50, RR 2.38 CI 35), but risks of akathisia, akinesia 0.1 to 55.7). and dystonia were equivocal. The- No study reported service out- re were no significant differences comes such as discharge from hos- between people given placebo and pital, or levels of satisfaction and those allocated fluphenazine in the quality of life, nor could we identify frequency of complaints of gastro- any direct economic evaluation of intestinal distress, cardiovascular fluphenazine.

14 Revista Colombiana de Psiquiatría, vol. XXXVI / No. 1 / 2007 Oral fluphenazine versus placebo for schizophrenia: a Cochrane systematic review of 40 years...

Discussion comparing oral fluphenazine with placebo has passed. It is surprising that there are so few trial-based data for the absolute Conclusions effects of fluphenazine. It is feasible that we have failed to identify some This review includes studies trials, but we think it unlikely that that span nearly four decades of we have missed any large studies. evaluative trials within psychiatry. Fluphenazine may well be antipsy- There is some empirical evidence chotic, but data in this review are that the quality of schizophrenia not convincing. Even in the short trial reporting has not changed over term, it is a drug prone to cause a time (27). We have found no time- variety of extrapyramidal and an- related differences in reporting of ticholinergic effects. Fluphenazine studies within this review and no is widely available and inexpensive suggestion of a change of effect sizes and for that reason alone it is un- across the decades. derstandable that it remains one of The seven included studies the many drugs used for treating in this review include people with people with serious mental illnes- schizophrenia who would be recog- ses. However, with weak evidence nisable in everyday practice. There for its positive effects and some ad- are those with strictly diagnosed verse effects that could be expensive illnesses, very likely to suffer from in terms of human suffering and schizophrenia, and people whose cost of treatment, it could prove illness was diagnosed using less better to use another inexpensive rigorous criteria. The dose of flu- drug supported by more favourable phenazine in the studies included data (25-26). in this review could be considered Even though this drug has been standard (mean 8.2 mg/day SD used as an antipsychotic drug for 3.9). Although the outcomes that decades, the fluphenazine story is have been used in this review are incomplete. Questions remain regar- accessible to both clinicians and ding the effect of this drug on global recipients of care, generalising to and mental state, quality of life and treatment in community settings, satisfaction and the consequences may be problematic. of the adverse effects. One or more The strength of this review is large, methodologically sound ran- that it presents up-to-date quanti- domised, placebo-controlled trials tative data for a benchmark treat- could help answer these questions. ment for schizophrenia which is However, with the advent of widely considered one of the essential available moderately effective antip- antipsychotics. Data, however, were sychotic drugs, the day for studies often inadequately reported and this

Revista Colombiana de Psiquiatría, vol. XXXVI / No. 1 / 2007 15 El-Din Matar H., Almerie MQ., Giraldo A., Adams C. rendered many outcomes unusable. tors. Brain research. Molecular brain Most trials report only six to twelve research. 1997;49(1-2):197-210. 6. BNF 48 (British National Formulary) by week outcomes for an illness that is the British Medical Association. 2004: mostly life-long. No studies reported p181 on service utilisation, economic 7. Deng MZ, Chen GQ, Phillips MR. Neuroleptic malignant syndrome in outcomes, or on satisfaction with 12 of 9,792 Chinese inpatients ex- care. It is not for us to judge past posed to neuroleptics: a prospective recommendations by standards of study. American Journal of Psychiatry. 1990;147(9):1149-1155 today. Now, however, it would seem 8. Essential Medicines. WHO Model List. prudent for WHO’s essential list 14th Edition. http://whqlibdoc.who. of antipsychotics to be revised to int/hq/2005/a87017_eng.pdf. include equally inexpensive and po- 9. Matar HE, AlMerie MQ. Oral fluphenazine versus placebo for schizophrenia. tentially accessible drugs, but with Cochrane Database of Systematic Re- better data on positive outcomes views 2007, Issue 1. Art. No.: CD006352. and more gentle proﬁles of adverse DOI: 10.1002/14651858.CD006352. 10. Higgins JPT, Green S (editors). Co- effects (26-27) chrane Handbook for Systematic Re- views of Interventions 4.2.5 (updated Acknowledgements May 2005). The Cochrane Library. 2005;(3). Thanks to Judith Wright (Leeds, 11. Schulz KF, Chalmers I, Hayes RJ, UK) for doing the electronic search, Altman DG. Empirical evidence of bias: dimensions of methodological Tessa Grant (Leeds, UK) Cochrane quality associated with estimates of Schizophrenia Group coordinator. treatment effects in controlled trials. JAMA. 1995;273:408-12. 12. DerSimonian R, Laird N: Meta-analysis References in clinical trials. Control Clin Trials. 1986, 7: 177-188. 1. Dencker SJ, Lepp M, Malm U. Do 13. Cook RJ, Sackett DL: The number nee- schizophrenics well adapted in the ded to treat: a clinically useful measure community need neuroleptics? A de- of treatment effect. BMJ. 1995, 310: pot neuroleptic withdrawal study. 452-454. 2. Kane JM, Woerner M, Sarantakos S. 14. Higgins JP, Thompson SG, Deeks JJ, Depot neuroleptics: A comparison Altman DG. Measuring inconsistency review of standard, intermediate and in meta-analyses. BMJ. 2003;327: low-dose regimens. Journal of Clinical 557-60. Psychiatry. 1986;47(suppl 5):30-3. 15. Altman DG, Bland JM: Detecting skew- 3. Darling HF. Fluphenazine: a preliminary ness from summary information. BMJ. study. Disease of the Nervous System. 1996, 313: 1200. 1959;20(4):167-70. 16. Hordern A, King A, Holt NF, Collins J, 4. Holt JP, Wright ER. Preliminary results Toussaint J. Thioproperazine in chro- with fluphenazine (prolixin) in chronic nic schizophrenia. British Journal of psychotic patients. American Journal Psychiatry. 1964;110:531-9. of Psychiatry. 1960;117:157-9. 17. Clark ML, Huber WK, Charalam- 5. Coirini H, Kallstrom L, Wiesel FA, Jo- pous KD, Serafetinides EA, Trousdale hnson AE. Modulation of basal ganglia W, Colmore JP. Drug treatment in neurotransmission by the classical newly admitted schizophrenic pa- antipsychotic fluphenazine due in part tients. Archives of General Psychiatry. to the blockade of dopamine D1-recep- 1971;25(5):404-9.

16 Revista Colombiana de Psiquiatría, vol. XXXVI / No. 1 / 2007 Oral fluphenazine versus placebo for schizophrenia: a Cochrane systematic review of 40 years...

18. Millar J. A trial of fluphenazine in schi- 23. Rifkin A, Quitkin F, Klein DF. Fluphe- zophrenia. British Journal of Psychia- nazine decanoate, oral fluphenazine, try. 1963;109:428-32. and placebo in treatment of remit- 19. Carpenter WT Jr, Buchanan RW, Kir- ted schizophrenics. II. Rating scale kpatrick B, Lann HD, Breier AF, Sum- data. Archives of General Psychiatry merfelt AT. Comparative effectiveness 1977;34(10):15-9. of fluphenazine decanoate injections 24. Rifkin A, Quitkin F, Rabiner CJ, Klein every 2 weeks versus every 6 wee- DF. Comparison of fluphenazine deca- ks. American Journal of Psychiatry. noate, oral fluphenazine, and placebo 1999;156(3):412-8. in remitted outpatient schizophre- 20. Egger M, Davey Smith G, Schneider nics. Psychopharmacology Bulletin. M, Minder C: Bias in meta-analysis 1976;12(2):24-6. detected by a simple, graphical test. 25. Soares BG, Fenton M, Chue P. Sulpiri- BMJ. 1997, 315: 629-634. de for schizophrenia. Cochrane Data- 21. Rifkin A, Quitkin F, Kane J, Klein DF, base Syst Rev. 2000;CD001162.) Ross D. The effect of fluphenazine 26. Rifkin A, Quitkin F, Rabiner CJ, Klein upon social and vocational functioning DF. Fluphenazine decanoate, fluphe- in remitted schizophrenics. Biological nazine hydrochloride given orally, and Psychiatry 1979;14(3):499-508. placebo in remitted schizophrenics. I. 22. Rifkin A, Quitkin F, Kane J, Klein DF. Relapse rates after one year. Arch Gen Fluphenazine decanoate, oral fluphe- Psychiatry. 1977;34(1):43-7. nazine, and placebo in the treatment 27. Thornley B, Adams C. Content and of remitted schizophrenics. II. Rating quality of 2000 controlled trials in scale data. Psychopharmacology schizophrenia over 50 years. BMJ. Bulletin 1977;13(2):40-50. 1998;317:1181-4.

Received to evaluation: January 18, 2007 Approved to publish: February 26, 2007

Correspondence Mr Hosam El-Din Matar Faculty of Medicine, Damascus University Damascus, Syria E-mail: [email protected]

Revista Colombiana de Psiquiatría, vol. XXXVI / No. 1 / 2007 17