
	Version PDF

A prospective study of the responsiveness of depression and suicidal ideation in acne patients to different phases of isotretinoin therapy

European Journal of Dermatology. Volume 15, Number 6, 484-8, November-December 2005, Clinical report

Summary

Author(s) : SC Kellett, DJ Gawkrodger , Department of Clinical Psychology, Keresforth Centre, Barnsley S70 6RS, UK., Department of Dermatology, Royal Hallamshire Hospital, Sheffield, S10 2JF, UK.

Summary : The degree and manner to which isotretinoin affects the mental health of acne patients has not been extensively researched, despite reports of possible associations between isotretinoin and depression. In this study, 33 patients with acne were prescribed a standard 16-week treatment regime with isotretinoin. At the initiation of isotretinoin treatment, week 8 review and termination of treatment, patients completed validated measures of depression, hopelessness and self-rated dermatological severity and were also assessed dermatologically using an acne grading protocol. When the first phase of isotretinoin treatment (week 1 to week 8) was compared to the second phase (week 9 to week 16), patients reported significant improvements in the cognitive-affective features of depression during the first phase of treatment, but not during the second phase. Corresponding improvements in the somatic symptoms of depression and hopelessness were not found. The implications of the research are discussed in relation to methodological design issues in this area.

Keywords : acne, depression, isotretinoin, suicidal ideation

ARTICLE

Auteur(s) : SC Kellett¹, DJ Gawkrodger²

¹Department of Clinical Psychology, Keresforth Centre, Barnsley S70 6RS, UK.
²Department of Dermatology, Royal Hallamshire Hospital, Sheffield, S10 2JF, UK

accepté le 1 Septembre 2005

Isotretinoin (13-cis-retinoic acid) was introduced in 1982 for the treatment of nodular and cystic acne, and more recently additionally in the treatment of less severe acne that has been found to be unresponsive to antibiotics [1]. Acne remains the most common reason for referral to specialist dermatological services [2]. Isotretinoin’s mechanism of action is fourfold; the reduction of sebum excretion [3], reduction of comedone formation [4], reduction of cutaneous and ductal Propionibactatium acnes [5] and reduction of general inflammation [6]. Isotretinoin is therefore active with regards to the four central aetiological factors involved in acne [7], which accounts for both its high degree of treatment efficacy [8] and presumably its increased usage in recent years [9]. Although the dermatological efficacy of isotretinoin is undoubted, the treatment is not without physical complications; isotretinoin is teratogenic and therefore contraindicated in pregnancy and can produce a range of unpleasant side-effects including commonly dry skin, lethargy, photosensitivity and abnormal liver and lipid functioning [10]. A less widely researched and far more controversial issue is the degree to which isotretinoin can create psychiatric problems in dermatology patients; including depression, psychotic symptoms, suicidal ideation and attempted suicide [11]. In terms of a possible mode of neuropsychological action, animal studies have illustrated that retinoids can cross into the central nervous system and may alter dopamine signalling systems important in mood regulation efforts [12]. Although intuitively appealing, human evidence regarding such neuropsychological pathways has not yet been forthcoming.The evidence regarding the association between psychiatric problems and isotretinoin usage [1] falls into four main themes; (a) case reports (b) reviews (c) systematic studies and (d) predictive studies. A number of case studies report patients (with no family history of depression) developing depression shortly after beginning isotretinoin [13, 14] or committing suicide on completing isotretinoin treatment [15, 16]. Obviously, such case reports are open to methodological criticism and the conclusions drawn therefore need to be necessarily conservative and reserved. Some case reports however include withdrawal/recommencement/increase of isotretinoin, which adds some degree of further methodological control and complexity. A case was reported of a male patient developing depression two weeks after starting isotretinoin, which initially responded to antidepressant medication, only to recur when the dose of isotretinoin was increased [17]. In a case series report, 24 patients were reported whose depression resolved once isotretinoin was withdrawn, only to recur again once the isotretinoin was recommenced [18].With regards to the reviews conducted, the methodologies have tended to focus on attempting to assess the proportion of patients developing depression once started on isotretinoin and subsequent psychiatric response to isotretinoin withdrawal. In a series of patients (n = 110) treated at doses between 1 and 2 mg/kg/day, 5.45% were found to develop depressive symptoms; with all such patients experiencing a remission in such symptoms once isotretinoin was stopped [19]. Spontaneous depression was reported to occur in 1% in a mixed dermatological diagnostic group (n = 700) being treated with isotretinoin, with depressive symptoms being seen to resolve within 2-7 days of discontinuation of isotretinoin [20]. In the US between the years 1982 and 2000, 37 patients committed suicide whilst taking isotretinoin, with a further 110 patients requiring in-patient psychiatric care. Of the patients committing suicide 22 per cent had a psychiatric history, whilst 44 per cent of the in-patient admissions had previous psychiatric diagnoses [21]. However, in a long-term physical safety study of isotretinoin usage [22], less than 0.5 of 1 per cent of patients were found to respond with depression, with the authors stating that such reactive depression was, on the balance of evidence, unlikely to be related to isotretinoin usage.In terms of systematic studies, there are a smaller number of such studies completed, which also unfortunately tend to be compromised by poor methodological design, with additional reliability and validity concerns regarding the psychiatric outcome indicators employed. In a prospective study of the side-effects associated with isotretinoin usage [10], depression was self-reported by 4% of 124 patients and tended to persist throughout treatment, but not with sufficient of intensity to prompt isotretinoin withdrawal. In a large sample size (n = 7535), retrospective cohort style study isotretinoin usage had no increase in relative risk estimates for depression, psychotic symptoms, suicide or attempted suicide [11]. Despite obvious adequate statistical power, the study suffered due to the inherent uncertainty regarding the validity of the diagnostic categories gleaned from computerised databases. A prospective study (n = 215) compared the mood and quality of life outcomes between isotretinoin and antibiotic/topical treatment groups [23]. During the course of isotretinoin treatment, group data indicated that patients with moderate levels of depression tended to remain stable over time (i.e. neither detriment nor improvement to reported mood state during treatment). Five patients were withdrawn from the study, due to apparent worsening of mood, two of whom agreed to be formally psychiatrically assessed. Such assessments tended to reveal sub-clinical depression, due to coexisting psychosocial stressors and patients were hence recommenced on isotretinoin without psychiatric incident.As is evident from the previous studies, attempts to analyse the possible association between that of isotretinoin and depression, have been hamstrung by poor methodological design including lack of randomised placebo controls, qualitative single case reports, small quantitative sample sizes, validity and reliability issues regarding measurement and limited data on clinical parameters. The relationship between acne and depression is further complicated by evidence that indicates that acne, due to its unsightly nature and subsequent likelihood to create body shame [24], tends possibly to predispose patients to developing depression. At present, it is effectively impossible to accurately forecast patients who may be prone to developing depression during or following isotretinoin treatment, with the pathway to a depressive presentation likely to be a multifactorial [1] and biopsychosocial phenomena.

Materials and methods

Participants

The sample was recruited by a consultant dermatologist (DJG) during routine clinical practice. All patients had been referred for specialist assessment and intervention due to non-responsiveness to standard acne treatment regimes in Primary Care. The study was approved by the South Sheffield Research Ethics Committee, with all patients providing written consent regarding their participation in the study. At assessment, the patients’ demographic details were collected alongside medical/drug histories and routine dematological and laboratory examinations were performed. The sample consisted of 21 males (63.63%) of an average age of 25.00 (sd = 6.67) and 12 females (36.36%) with an average age of 24.75 (sd = 4.71). All patients were assessed as suitable for treatment with isotretinoin (13-cis-retinoic acid; Roaccutane), which was subsequently prescribed at a standard dose of 1mg/kg for 16 weeks. Patients were dermatologically re-assessed after 8 weeks of treatment and on conclusion of isotretinoin treatment at 16 weeks. At each assessment point (i.e. initiation of treatment, mid-treatment at week 8 and termination of treatment at week 16), patients were requested to complete a range of psychological measures and dermatological self-ratings. 21 patients completed measures at week 8 and 22 completed measures at termination of treatment (week 16).

Psychological measures

At assessment, week 8 and week 16, patients completed both the 21 item Beck Depression Inventory (BDI; Beck, 25) and the 20 item Beck Hopelessness Scale (BHS; Beck, 26, 27). The BDI has been extensively examined and illustrated to have good validity and sound reliability [25]. The BDI is a valid and reliable measure of the intensity of depressive symptomatology and has been used extensively in research studies designed to assess the efficacy of pharmacological interventions and in the detection of depression in medical patients [28]. The BDI can be treated as a single total scale score or as two independent scales, which have been indicated via factor analysis. The first scale is titled ‘cognitive-affective’ and measures the intensity of the psychological symptoms of depression such sadness, guilt and worthlessness. The second scale is titled ‘somatic’ and measures the physical symptoms of depression such as appetite, libido and sleep difficulties [25]. On the total BDI, scores between 0-13 indicate minimal depression, 14-19 mild, 20-28 moderate and 29-63 severe [25]. The BHS has similarly been extensively studied and found to have satisfactory psychometric properties [26, 27]. A cut-off score of 9 or above on the BHS is predictive of suicide completers [27].

Dermatological ratings; clinician and patient ratings

The severity of acne was graded clinically by a Consultant Dermatologist (DJG) using a validated acne assessment tool [29]. This system essentially entails using clinical observations to grade the presenting acne on a 0-8 scale. Patients also completed a self-assessment of acne severity using a 0-8 visual analogue scale for their face, chest and back. This rating system has been previously utilised in isotretinoin evaluation research [30].

Statistical analysis

The total and sub-scale BDI, BHS and dermatological rating scores were examined using repeated measure ANOVAs, with post-hoc comparisons of mean difference performed to examine possible differentiation between the phases of isotretinoin treatment (i.e. week 8-16; phase 1, week 9-16; phase 2). For each scale, Mauchly’s Test of Sphericity was non-significant, indicating that data was suitable for ANOVA.

Results

Analysis of the differences on the scales between those who completed the study and those who dropped out of treatment revealed that those patients who completed treatment had significantly higher ‘cognitive-affective’ scores on the BDI at assessment (t = 2.05, sig. < 0.05). No other differences were apparent on any of the other psychological or dermatological scales between completers and non-completers. No gender differences on any of the psychological or dermatological scales were evident at any time during the course of isotretinoin treatment. The mean duration of the skin disease in the treatment completers was 6.60 (sd 4.08) years.

Table 1( Table 1 ) summarises the results from the repeated measures ANOVA on the psychological measures and the dermatological ratings over the course of the isotretinoin treatment and compares the BDI data with available comparable evidence [23]. In terms of the psychological scales, the ANOVA indicate that there was no difference over the course of treatment on overall depression scores, overall hopelessness and the somatic symptoms of depression. However there was a significant reduction over the course of isotretinoin treatment on the cognitive-affective scale of the BDI. Post hoc pair-wise comparisons of the phases of treatment indicate that there was a significant reduction in cognitive-affective features of depression between assessment and 8-week review (phase 1, mean difference = 2.76, p = 0.02), but not between week 8 and termination of treatment (phase 2, mean difference = – 0.47, p = 0.71). Comparison of the BDI data with that of the available evidence [23], illustrates higher depression score means at assessment, mid-term and completion of isotretinoin treatment in the current sample.

In terms of the dermatological ratings, both dermatologist and self-rated facial and back acne significantly reduced over the course of the isotretinoin treatment. The reductions in the severity of acne on the chest (self-rated), did not however reach statistical significance. In terms of the dermatologist rated severity of acne, there were significant improvements both between assessment and 8-week review (phase 1, mean difference = 2.14, p = 0.01) and week 9 and termination of treatment (phase 2, mean difference = 1.35, p =0.01). On the self-rated severity of face and back acne, significant improvements were also recorded in both phases of isotretinoin treatment. There were no significant correlations between dermatologist and self-rating of acne severity at any point during isotretinoin treatment.

Table 2( Table 2 ) contains the categories of the patients on the BDI as they progress through isotretinoin treatment. 6.45% (n = 2) were in the severely depressed category at the start of treatment (and admitting suicidal ideation), which fell to an n = 1 at week 8 and termination of treatment (and still admitting suicidal ideation). Examination of the BHS scores equal to or greater than 9 and therefore indicating suicide potential revealed that 9.09% (n = 3) of the patients at assessment were reporting significant feelings of hopelessness, which fell to a single patient at week 8 and at termination of treatment.
Table 1 Psychological and dermatological scores during isotretinoin treatment

	Assessment Mean (sd)	Mid-treatment Mean (sd)	Termination Mean (sd)	F-Value
BDI (Ng et al 2002)	3.5 (4.00)	3.1 (3.7)	3.0 (3.9)
BDI	13.29 (14.82)	10.35 (12.64)	10.11 (12.64)	1.84
BDI: Cognitive-Affective scale	9.17 (9.93)	6.41 (9.59)	5.94 (8.98)	3.33*
BDI: Somatic scale	3.58 (4.54)	3.35 (3.55)	3.58 (3.67)	0.09
BHS	4.76 (5.61)	4.58 (5.05)	3.88 (4.58)	0.19
Clinician acne grading score	3.92 (2.81)	1.78 (1.47)	0.42 (0.64)	14.14**
Self-rated; face	5.64 (1.90)	1.50 (1.01)	0.71 (0.91)	63.28**
Self-rated; chest	1.28 (1.72)	0.50 (0.65)	0.35 (1.33)	2.81
Self-rated; back	2.85 (2.82)	0.64 (0.92)	0.28 (0.82)	10.64**

Table 2 BDI categories during course of isotretinoin treatment

	Minimal depression (0-13 BDI score range)	Mild depression (14-19 BDI score range)	Moderate depression (20-28 BDI score range)	Severe depression (29-63 BDI score range)
Assessment (n = 33)	74.19%	9.67%	9.67%	6.45%
Week 8 (n = 21)	80.95%	4.76%	9.52%	4.76%
Termination (n = 22)	81.81%	4.54%	9.09%	4.54%

Discussion

The current study has attempted to address the association between isotretinoin treatment and depression via a prospective study. An obvious issue of debate and concern is the degree to which acne exerts a negative effect on mood, even prior to an attempt to intervene dermatologically. Patients may well have already ‘failed’ on antibiotic/topical treatment provided via Primary Care, in the context of experiencing ‘skin failure’ [31] and the acknowledged stigma of acne [24], thereby creating the necessary conditions for depression to occur in some psychologically vulnerable patients. In short, patients with severe and/or recalcitrant acne may well be hopeless prior to treatment with isotretinoin, especially given that the average duration of acne in the study was over six years. The provision of an antibiotic control group in future studies of isotretinoin would appear a useful methodological addition. The current study has used a validated measure of hopelessness, a measure that has not been employed previously in psychodermatological research. The BHS data indicated that 9.09% of the sample had sufficient levels of hopelessness prior to treatment, to warrant concern as to possible risk of suicide completion. The figure of approximately nine per cent is broadly similar to previous research investigating prevalence of suicidal ideation in dermatology patients [32]. Such previous research is open to criticism however due to employing single item assessments, which are notoriously, both clinically and psychometrically, unreliable. The single patient who remained hopeless at the mid-point and termination of isotretinoin treatment in the current study, experienced dermatological improvements in their acne indicating that the depression was probably unrelated to the skin disease. No patients were removed from the study due to concerns regarding their mental health.

The design of the current study facilitated the comparison of the phases of isotretinoin treatment in terms of dermatological and psychological functioning, an analysis that has not been previously attempted. Use of the ‘somatic’ and ‘cognitive-affective’ sub-scales of the BDI revealed results that were masked in the overall BDI score and may have been masked in previous research using the BDI as an outcome indicator [23]. The current data suggest that the significant improvements in the cognitive-affective features of depression recorded during treatment, took place specifically in the first eight weeks of treatment, but were not subsequently statistically maintained during the second phase of treatment. It is interesting to note that the group of patients that completed treatment scored significantly higher on the ‘cognitive-affective’ sub-scale at assessment, in comparison to people who chose to eventually terminate treatment. This may suggest that patients who are experiencing negative moods prior to treatment are more likely to remain in treatment and endure the likely side-effects of isotretinoin [10], in comparison to patients with higher moods at assessment. The methodology did not include investigation of the reasons for withdrawal from isotretinoin treatment and the study. No tandem changes in somatic features of depression were recorded in either of the phases of isotretinoin treatment. The lack of reduction in somatic features of depression over the course of treatment, may be due to BDI scale items such as the ‘loss of energy’ (BDI item 15) and ‘fatigue’ (BDI item 20) measuring aspects of lethargy, which are an acknowledged and common side-effect of isotretinoin treatment [10]. More specific research regarding the relationship and interplay between mood and the physical side-effects of isotretinoin treatment is urgently required. No gender differences were apparent across the treatment duration, which is unusual in psychodermatological research [24]. However, previous research has not tended to employ measures such as the BDI and BHS. Gender differences may well occur in aspects of social presentation [24], due to females tending to be more embarrassed by their skin diseases than males, yet the presence of such gender differences in dermatologically-related depression is under researched. The fact that the current study employed the BDI and the BHS has proved useful, as these well-validated measures have been previously under utilised in investigations of isotretinoin and psychodermatological issues in general. The means prior to, during and on completion of isotretinoin, on the BDI in the current study was elevated in comparison to previous research [23]. Patients in the comparison data were also treated with isotretinoin for 8 weeks longer, on average, than the current sample. It is difficult to interpret such results without additional sources of evidence on the BDI from dermatological populations.

The design of the current study ensures that the results need to be interpreted with a degree of caution, due to the acknowledged absence of a randomised control group. Open study designs such as the one employed are obviously prone to bias in sampling that may, as a result, prejudice ratings on outcome measures such as the BHS and BDI. The employment of a placebo control group in future research regarding isotretinion and depression is scientifically indicated, although ethically complex and troubling. The moral dilemma involved in withholding a dermatologically efficacious treatment in a group of patients with recalcitrant/severe acne, that may well respond with improvements in psychological functioning [30, 33, 34], are yet to be debated and decided. The study was also compromised by the small sample size. Due to the likelihood emerging from the available evidence-base that depression is a possible, but not probable, consequence of isotretinoin treatment in a proportion of patients, samples in future prospective studies need to be sizeable, in an attempt to capture the apparent low base-rate of depressive response to isotretinoin. The design of such studies would also need ideally to include some form of ‘no treatment, community control group’, in order to provide comparisons with the rates of depression associated with acne in the community. Additionally, in future research designs, once patients have been identified who have apparently reacted negatively to isotretinoin, then full psychological and psychiatric assessments should be completed for such patients. Such investigations would increase the understanding of the poorly researched issues of the role (or not) of prior psychiatric and psychological morbidity issues in sensitivity to isotretinoin therapy.

The design of future alternative methodologies would also benefit from far more regular sampling of mood and other psychological features than conducted in the present and previous investigations regarding isotretinoin. Such intensive sampling techniques may well produce more fine-grained analysis of the timing, frequency and duration of mood change during isotretinoin treatment and response to drug withdrawal/reintroduction. Such time sampling issues also need to take into consideration the period of time following isotretinoin treatment completion. The current study would have benefited from following patients up after treatment termination. The long-term consequences of isotretinoin treatment upon rates of subsequent possible mood disorders are under researched, as are questions regarding the maintenance of demonstrated positive psychological effects [30, 33, 34]. Although the current study has highlighted the previously unrecognised psychological importance of the first eight weeks of isotretinoin treatment, the data do not support the presence of a direct and causal relationship regarding isotretinoin and depression in the patients studied. Isotretinoin still remains somewhat of a psychiatric enigma, and will continue to remain so, until adequately sized and adequately controlled research designs become available.

Acknowledgements

The authors identify no conflicts of interest and have received no financial support for this study from any external agency.

References

1 Hong C, Schweitzer I. The association between depression and isotretinoin usage in acne. Austr New Zeal J Psych 2003; 37: 78-84.

2 Pawin H, Beylot C, Chivot M, Fuare M, Poli F, Reuuz J, et al. Physiopathology of acne vulgaris: recent data, new understanding of treatments. Eur J Dermatol 2004; 14: 4-12.

3 King K, Jones DH, Daltrey DC, et al. A double-blind study of the effect of 13-cis-retinoic acid on acne, sebum excretion rate and microbial population. Br J Dermatol 1982; 107: 583-90.

4 Daiziel K, Barton S, Marks R. The effects of isotretinoin on fbilicular and sebaceous gland differentiation. Br J Dermatol 1987; 117: 317-23.

5 Leyden JJ, McGinfey KJ. Effect of 13-cis-retinoic acid on sebum production and propionibacterium acnes in severe nodulocystic acne. Arch Dermatol Res 1982; 272: 331-7.

6 Falton RH, Lee WL, Shalita AR, et al. In vivo effect of isotretinoin on monocyte chemotaxis. J Invest Dermatol 1986; 86: 550-2.

7 Cunliffe WJ, Baxter K. The place of oral isotretinoin in the treatment of acne: Benefits and side effects. CME Bulletin Dermatol 2000; 2: 56-62.

8 Farrell LN, Strauss JS, Stranieri AM. The treatment of severe cystic acne with 13-cis-retinoic acid: Evaluation of sebum production and the clinical response in a multiple – dose trial. J Am Acad Dermatol 1980; 3: 602-11.

9 Dreno B, Daniel F, Allaert FA, Aube I. Acne; evolution of the clinical practice and therapeutic management of acne between 1996 and 2000. Eur J Dermatol 2003; 13: 166-70.

10 Hull PR, Demkiw-Bartel C. Isotretinoin use in acne: Prospective evaluation of adverse effects. J Cuta Med Surg 2000; 4: 66-70.

11 Jick SS, Kremers HM, Vasilakis-Scaramozza C. Isotretinoin use and risk of depression, psychotic symptoms suicide and attempted suicide. Arch Dermatol 2000; 136: 1231-6.

12 Krezel W, Ghyselinck N, Samad TA, et al. Impaired locomotion and dopamine signalling in retinoid receptor mutant mice. Science 1998; 279: 863-7.

13 Byne A, Hnatko G. Depression associated with isotretinoin therapy. Canadian J Psych 1995; 40: 567.

14 Duke E, Guenther L. Psychiatric reactions to the retinoids. Can J Dermatol 1993; 5: 467.

15 Bravard P, Krug M, Rzeznick JC. Isotretinoin et depression, soyons vigilantes. Nouv Dermatol 1993; 12: 15.

16 Gatti S, Serri F. Acute depression from isotretinoin. J Am Acad Dermatol 1991; 25: 132.

17 Ng CH, Tam MM, Hook SJ. Acne, isotretinoin treatment and acute depression. W J Biol Psych 2001; 2: 159-61.

18 Ault A. Isotretinoin use may be linked with depression. Lancet 1998; 351: 730.

19 Hazen PG, Carney JF, Walker AE, Stewart JJ. Depression – a side effect of 13-cis-retinoic acid. J Am Acad Dermatol 1983; 9: 278-9.

20 Scheinman PL, Peck GL, Rubinow DR, DiGiovanna JJ, Abangan DL, Ravin PD. Acute depression and isotretinoin. J Am Acad Dermatol 1990; 22: 1112-4.

21 Wysowski DK, Pitts M, Beitz J. Depression and suicide in patients treated with isotretinoin. N Engl J Med 2001; 344: 460.

22 Goulden V, Layton AM, Cunliffe WJ. Long-term safety of isotretinoin in the treatment of acne vulgaris. Br J Dermatol 1994; 131: 360-3.

23 Ng CH, Tam MM, Celi E, Tate B, Schweitzer I. A prospective study of depressive symptoms and quality of life in acne vulgaris patients treated with isotretinoin compared to antibiotic and topical therapy. Aust J Dermatol 2002; 43: 262-8.

24 Kellett SC. Shame-fused acne: A biopsychosocial conceptualisation and treatment rationale. In: Gilbert P, Miles J, eds. Body Shame: Conceptualisation, Research and Treatment. Hove: Brunner-Routledge, 2002.

25 Beck AT, Steer RA, Brown GK. BDI-II manual. The Psychological Corporation. San Antonio: Harcourt Brace & Company, 2000.

26 Beck AT, Kovacs M, Weissman A. Hopelessness and suicidal behaviour: An overview. JAMA 1975; 234: 1146-9.

27 Beck AT, Brown G, Berchick RJ, Stewart BL, Steer RA. Relationship between hopelessness and ultimate suicide: A replication with psychiatric out-patients. Am J Psych 1990; 147: 190-5.

28 Zich JM, Attkisson CC, Greenfield TK. Screening for depression in Primary Care: The CES-D and the BDI. Intern J Psych Med 1990; 20: 259-77.

29 Allen BS, Smith JG. Various parameters for grading acne vulgaris. Arch Dermatol 1982; 118: 23-7.

30 Kellett SC, Gawrodger DJ. The psychological and emotional impact of acne and the effect of treatment with isotretinoin. Br J Dermatol 1999; 140: 273-82.

31 Cotterill JA, Cunliffe WJ. Suicide in dermatology patients. Br J Dermatol 1988; 137: 246-50.

32 Gupta MA, Gupta AK. Depression and suicidal ideation in dermatology patients with acne, alopecia areata, atopic dermatitis and psoriasis. Br J Dermatol 1998; 139: 846-50.

33 Rubinow DR, Peck GL, Squillance KM, Gantt GG. Reduced anxiety and depression in cystic acne patients after successful treatment with isotretinoin. J Am Acad Dermatol 1987; 17: 25-32.

34 MacDonald Hull S, Cunliffe WJ, Hughes BR. Treatment of the depressed and dysmorphic acne patient. Clin Exp Dermatol 1991; 16: 210-1.