Effects of Oral Isotretinoin on Lipids and

Liver Enzymes in Acne Patients

Okan Kızılyel, MD; Mahmut Sami Metin, MD; Ömer Faruk Elmas, MD; Yasemin Çayır, MD; Akın Aktas, MD

Practice Points
 Isotretinoin is recommended for treatment of severe inflammatory acne and for cases resistant to prior
treatment with antibiotics or topical agents; however, it may cause alterations in lipids and liver enzymes.
 In our study, liver enzymes were less affected than lipids in patients who were treated with isotretinoin.

y
 Use caution when administering isotretinoin in patients with a history of abnormal findings.

op
tc
A
Isotretinoin has been used to treat severe inflam- cne is a chronic inflammatory condition of
matory acne that is resistant to antibiotics or topi- the pilosebaceous unit affecting approxi-
no
cal agents; however, it also may cause alterations mately 79% to 95% of adolescents in the
in lipids and liver enzymes. In this retrospective Western world.1 Treatment of acne depends on
study, we evaluated changes in lipids and liver its severity. Topical tretinoin, adapalene, benzoyl
o

enzymes in 322 acne patients who had been peroxide, azelaic acid, and topical antibiotics gen-
treated with oral isotretinoin at our institution over a erally are used in cases of noninflammatory or
D

3-year period. Each patient’s medical records were mild inflammatory disease. Isotretinoin is recom-
evaluated to determine baseline triglyceride (TG), mended for treatment of severe inflammatory acne
S

low-density lipoprotein (LDL), high-density lipopro- (eg, nodulocystic or conglobata acne) and for cases
TI

tein (HDL), aspartate aminotransferase (AST), and of acne that have proven to be resistant to prior
alanine aminotransferase (ALT) levels compared to treatment with antibiotics or topical agents. Dosages
of isotretinoin range from 0.5 to 2 mg/kg daily for
U

levels recorded at 3 and 6 months following initia-

tion of treatment with oral isotretinoin. Overall, sta- 16 to 24 weeks.1 Isotretinoin reduces the activity
C

tistically significant increases in TG and LDL levels and size of the sebaceous glands, normalizes kera-
were noted following treatment with isotretinoin tinization of the sebaceous follicles, and decreases
(P.001 and P.001, respectively), while HDL lev- the number of Propionibacterium acnes. Isotretinoin
els were shown to decrease (P.001). Statistically also may cause clinical side effects and laboratory
significant increases in AST levels also were noted changes, the most important being teratogenicity. It
(P.016). Although ALT levels also increased, the also may cause mucocutaneous side effects including
changes were not statistically significant (P.72). cracked lips, dryness of the skin and nasal mucosa,
In our study, isotretinoin appeared to have a greater skin redness, eye dryness, and eye irritation.1 It
effect on lipids than liver enzymes. Dermatologists also may cause blepharoconjuctivitis, photosensi-
should not avoid isotretinoin use for appropriate tivity, asteatotic dermatitis, pruritus, telogen efflu-
indications, but close follow-up is important. vium, secondary bacterial colonization, nail fragility,
Cutis. 2014;94:234-238. periungual pyogenic granuloma, paronychia, myal-
gia, intracranial hypertension, nausea, headache,
From the Faculty of Medicine, Atatürk University, Erzurum, Turkey. vomiting, depression, psychosis, suicide, constipa-
Drs. Kızılyel, Metin, Elmas, and Aktas are from the Department of tion, and allergic reactions.2 Isotretinoin treatment
Dermatology. Dr. Çayır is from the Department of Family Medicine.
may increase serum levels of liver enzymes, triglyc-
The authors report no conflict of interest.
Correspondence: Okan Kızılyel, MD, Department of Dermatology,
erides (TGs), and low-density lipoprotein (LDL)
Faculty of Medicine, Atatürk University, 25000 Erzurum, Turkey cholesterol, and reduce the level of high-density
([email protected]). lipoprotein (HDL) cholesterol.1 This retrospective

234 CUTIS® WWW.CUTIS.COM

Copyright Cutis 2014. No part of this publication may be reproduced, stored, or transmitted without the prior written permission of the Publisher.
 Effects of Oral Isotretinoin

study sought to evaluate the effect of isotretinoin on (standard deviation [SD]) age of the women was 23.9
liver enzymes and lipids over 6 months. (5.4) years and the mean (SD) age of the men was 23.8
(7.02) years. Most of the patients were on a regimen of
Materials and Methods 30 or 40 mg of isotretinoin daily. Differences between
Our retrospective study was conducted at the Hospital dosages and laboratory values were not analyzed because
of Atatürk University in Erzurum, a city located in east- we assumed there would not be a significant differ-
ern Turkey. All patients who were treated in the depart- ence, as most patients received the same dose. The
ment of dermatology and had received oral isotretinoin mean (SD) WBC was 8.4 (3.5)103/mL. The mean
between June 2009 and June 2012 were included in (SD) RBC was 4.9 (0.5)106/mL. The mean (SD)
the study. The study was based on an evaluation of hemoglobin count was 14.3 (1.7)106/mL (women,
the patients’ medical records. All patients received 13.6 [1.5]106/mL; men, 15.9 [1.1]106/mL).
oral isotretinoin 0.5 to 1 mg/kg daily; the majority of The study evaluated the effects of isotretinoin
patients received 30 to 40 mg daily. Patient medical on liver enzymes (AST and ALT) and lipids (TGs,
records included age; gender; white blood cell (WBC) LDL, and HDL). Nearly all of the patients (95%)
count; red blood cell (RBC) count; hemoglobin count; had normal AST and ALT levels at baseline. The
and aspartate aminotransferase (AST), alanine amino- results are outlined in the Table. Some values were
transferase (ALT), TG, LDL, and HDL levels at the not recorded for all patients at each follow-up.

y
beginning of treatment. Aspartate aminotransferase, Aspartate Aminotransferase Analysis—Aspartate

op
ALT, TG, LDL, and HDL levels also were measured aminotransferase levels were classified as normal
at 3- and 6-month follow-up. Analysis of AST, ALT, and high. At baseline, mean (SD) AST levels
TG, LDL, and HDL levels was based on the National were 20.2 (6.6) U/L, with normal levels in 311
Cholesterol Education Program guidelines.3 Aspartate
aminotransferase and ALT levels were classified as
tc (96.6%) patients and high in 7 (2.2%) patients. At
3-month follow-up, mean (SD) AST levels were
normal (40 U/L) and high (≥40 U/L). Triglyceride 20.7 (5.2) U/L, with normal levels in 270 (83.9%)
no
levels were classified as normal (150 mg/dL), border- patients and high levels in 3 (0.9%) patients. At
line high (150–199 mg/dL), high (200–499 mg/dL), 6-month follow-up, mean (SD) AST levels were
and very high (≥500 mg/dL). Low-density lipoprotein 21.3 (5.7) U/L, with normal levels in 209 (64.9%)
o

levels were classified as optimal (100 mg/dL), above patients and high levels in 4 (1.2%) patients.
optimal (100–129 mg/dL), borderline high Aspartate aminotransferase levels increased at 3- and
D

(130–159 mg/dL), high (160–189 mg/dL), and very 6-month follow-up compared to baseline. Differences
high (≥190 mg/dL). High-density lipoprotein levels between AST levels were statistically significant
S

were classified as low (40 mg/dL), normal (F2,4164.2, P.016). Differences between AST lev-
(40–59 mg/dL), and high (≥60 mg/dL). Normal WBC
TI

els at baseline and 3-month follow-up were not

was defined as 3.5 to 12.5103/mL. Normal hemo- statistically significant (P.3). Differences between
globin count was defined as 11.5 to 15.0106/mL for AST levels at 3- and 6-month follow-up were not
U

women and 13 to 17106/mL for men. Normal RBC statistically significant (P.4). Differences between
C

was defined as 4.0 to 5.2106/mL for women and 4.5 to AST levels at baseline and 6-month follow-up were
5.9106/mL for men. Statistical analysis was performed statistically significant (P.07). Differences between
using SPSS version 17.0. Repeated measures analysis of AST classifications at the 3 time points were not
variance was used to compare means between 3 groups statistically significant (F2,4160.44, P.64). Overall,
(baseline, 3-month, and 6-month values). A paired the results indicated that AST levels increased over
sample t test was used to compare means between time in patients treated with isotretinoin, but the
any 2 groups. Results with P.05 were considered sta- increase was not above the normal range and was not
tistically significant. statistically significant.
Alanine Aminotransferase Analysis—Alanine ami-
Results notransferase levels were classified as normal or
Treatment with oral isotretinoin was initiated in high. At baseline, mean (SD) ALT levels were
349 patients at our institution from June 2009 to 16.8 (11.2) U/L, with normal levels in 303 (94.1%)
June 2012. Twenty-seven of these patients were patients and high in 19 (5.9%) patients. At
excluded from the study because their medical 3-month follow-up, mean (SD) ALT levels were
records were not available. Medical records from 16.2 (9.3) U/L, with normal levels in 263 (81.7%)
322 patients were obtained. The study popula- patients and high in 11 (3.4%) patients. At
tion consisted of 226 (70.2%) women and 96 6-month follow-up, mean (SD) ALT levels were
(29.8%) men. Patients ranged in age from 17 to 17.0 (11.3) U/L, with normal levels in 201 (62.4%)
64 years, with a mean age of 23.9 years. The mean patients and high in 11 (3.4%) patients. Alanine

WWW.CUTIS.COM VOLUME 94, NOVEMBER 2014 235

Copyright Cutis 2014. No part of this publication may be reproduced, stored, or transmitted without the prior written permission of the Publisher.
 Effects of Oral Isotretinoin

Summary of Resultsa
Patients, n (%)
Laboratory Value Baseline 3 Months 6 Months F Score, P Value
AST F2,4160.44, P.64
Normal (40 U/L) 311 (96.6) 270 (83.9) 209 (64.9)
High (≥40 U/L) 7 (2.2) 3 (0.9) 4 (1.2)
ALT F2,4180.21, P.54
Normal (40 U/L) 303 (94.1) 263 (81.7) 201 (62.4)
High (≥40 U/L) 19 (5.9) 11 (3.4) 11 (3.4)
TG F2,3866.9, P.001b
Normal (150 mg/dL) 270 (83.9) 197 (61.2) 145 (45)
Borderline high (150–199 mg/dL) 30 (9.3) 38 (11.8) 36 (11.2)
High (200–499 mg/dL) 20 (6.2) 22 (6.8) 16 (5)

y
Very high (≥500 mg/dL) 2 (0.6) 1 (0.3) 0 (0)

op
LDL F2,38251.2, P.001b
Optimal (100 mg/dL) 162 (50.3) 89 (27.6) 60 (18.6)
Above optimal (100–129 mg/dL)
Borderline high (130–159 mg/dL)
95 (29.5)
32 (9.9)
tc98 (30.4)
54 (16.8)
84 (26.1)
44 (13.7)
no
High (160–189 mg/dL) 11 (3.4) 12 (3.7) 8 (2.5)
Very high (≥190 mg/dL) 3 (0.9) 5 (1.6) 1 (0.3)

HDL F2,3845.2, P.006b

o

Low (40 mg/dL) 60 (18.6) 63 (19.6) 48 (14.9)

Normal (40–59 mg/dL) 173 (53.7) 154 (47.8) 117 (36.3)
D

High (≥60 mg/dL) 71 (22) 41 (12.7) 33 (10.2)

S

Abbreviations: AST, aspartate aminotransferase; ALT, alanine aminotransferase, TG, triglyceride; LDL, low-density lipoprotein;
HDL, high-density lipoprotein.
TI

a
Some values were not recorded for all patients at each follow-up.
b
Statistically significant.
U
C

aminotransferase levels at 3-month follow-up in 20 (6.2%) patients, and very high in 2 (0.6%)
were lower than baseline but higher at 6-month patients. At 3-month follow-up, mean (SD) TG
follow-up compared to baseline and 3-month levels were 117 (60) mg/dL, with normal levels in
follow-up. Overall, ALT levels increased with time, 197 (61.2%) patients, borderline high in 38 (11.8%)
but the differences between baseline and 3- and patients, high in 22 (6.8%) patients, and very high
6-month follow-up were not statistically signifi- in 1 (0.3%) patient. At 6-month follow-up, mean
cant (F2,4160.32, P.72). Differences between ALT (SD) TG levels were 122 (65) mg/dL, with normal
classifications at each time point were not statisti- levels in 145 (45%) patients, borderline high in
cally significant (F2,4180.21, P.54). Overall, the 36 (11.2%) patients, high in 16 (5%) patients, and
results indicated that ALT levels increased over time very high in 0 (0%) patients. Triglyceride levels
in patients treated with isotretinoin, but the increase increased and differences between TG levels at base-
was not statistically significant. line and 3- and 6-month follow-up were statistically
Triglycerides Analysis—Triglyceride levels were significant (F2,38417, P.001). Baseline TG levels
classified as normal, borderline high, high, and compared to 3-month follow-up were statistically signif-
very high. At baseline, mean (SD) TG levels were icant (P.001). Differences in TG levels at 6-month
107 (71) mg/dL, with normal levels in 270 (83.9%) follow-up versus baseline were statistically signifi-
patients, borderline high in 30 (9.3%) patients, high cant (P.001). However, changes in TG levels from

236 CUTIS® WWW.CUTIS.COM

Copyright Cutis 2014. No part of this publication may be reproduced, stored, or transmitted without the prior written permission of the Publisher.
 Effects of Oral Isotretinoin

3- to 6-month follow-up were not statistically signif- decreases were noted in HDL levels (F2,38419,
icant (P.21). Differences between TG classifica- P.001). Differences between baseline HDL levels
tions at each time point were statistically significant compared to 3-month follow-up were statistically
(F2,3866.9, P.001). Overall, TG levels increased significant (P.001). Differences between baseline
from baseline during isotretinoin treatment at 3- and HDL levels compared to 6-month follow-up were
6-month follow-up, and these increases were above statistically significant (P.001). Differences in
normal range; however, there was no statistically HDL levels at 3- and 6-month follow-up were sta-
significant increase from 3- to 6-month follow-up. tistically significant (P.001). Differences between
Low-Density Lipoprotein Analysis—Low-density HDL classifications at each time point were statisti-
lipoprotein levels were classified as optimal, above cally significant (F2,3845.2, P.006). Overall, there
optimal, borderline high, high, and very high. were statistically significant decreases in HDL levels
At baseline, mean (SD) LDL levels were during isotretinoin treatment from baseline and this
102 (28) mg/dL, with optimal levels in 162 (50.3%) decrease was above normal range; however, HDL
patients, above optimal in 95 (29.5%) patients, levels did not decrease at 3- and 6-month follow-up.
borderline high in 32 (9.9%) patients, high in
11 (3.4%) patients, and very high in Comment
3 (0.9%) patients. At 3-month follow-up, mean Studies in the literature evaluating the effects of

y
(SD) LDL levels were 113 (30) mg/dL, with opti- isotretinoin on liver enzymes and lipids suggested

op
mal levels in 89 (27.6%) patients, above optimal in that oral isotretinoin may cause alterations in liver
98 (30.4%) patients, borderline high in 54 (16.8%) aminotransferases (AST and ALT), TGs, HDL,
patients, high in 12 (3.7%) patients, and very high and LDL in various degrees.1 Zane et al4 studied
in 5 (1.6%) patients. At 6-month follow-up, mean
(SD) LDL levels were 113 (27) mg/dL, with opti-
tc 13,772 patients with acne undergoing oral isotretinoin
therapy between March 1995 and September 2002.
mal levels in 60 (18.6%) patients, above optimal in The investigators found increased liver transaminase
no
84 (26.1%) patients, borderline high in 44 (13.7%) and serum lipid levels. They suggested that these
patients, high in 8 (2.5%) patients, and very high abnormalities were generally transient and reversible.4
in 1 (0.3%) patient. Overall, there were statistically Bershad et al5 reported an increase in LDL and TG but
o

significant increases in LDL levels at 3- and 6-month a decrease in HDL during isotretinoin therapy. These
follow-up (F2,38275, P.001). Differences between changes in the lipid profile also appeared to be tran-
D

baseline LDL levels and 3-month follow-up were sta- sient and returned to baseline level 2 months follow-
tistically significant (P.001). Differences between ing the end of treatment.5 In another study of
S

baseline LDL levels and 6-month follow-up were sta- 130 patients who were treated with isotretinoin,
TI

tistically significant (P.001). However, differences Vieira et al1 noted an increase in AST, ALT, and
in LDL levels at 3- and 6-month follow-up were TG levels. Most of the studies in the literature that
not statistically significant (P.74). Differences reported effects of isotretinoin on liver enzymes and
U

between LDL classifications at each time point lipids suggested that the effects were reversible.
C

were statistically significant (F2,38251.2, P.001). Although many studies reported alterations in
Overall, statistically significant increases in LDL serum transaminase and lipid levels, other studies
levels from baseline were noted during isotretinoin reported no effect. In one study of 150 participants,
treatment and this increase was above normal range; Brito et al2 found no statistically significant changes
however, LDL levels did not significantly increase in liver transaminase, TG, HDL, or LDL levels
from 3- to 6-month follow-up. following treatment with isotretinoin. In another
High-Density Lipoprotein Analysis—High-density study of 1292 participants by Alcalay et al,6 serum
lipoprotein levels were classified as low, normal, levels of liver enzymes were not elevated to a degree
and high. At baseline, mean (SD) HDL levels were necessitating discontinuation of isotretinoin treat-
52.4 (16) mg/dL, with low levels in 60 (18.6%) ment. In another study of 30 participants, Baxter
patients, normal in 173 (53.7%) patients, and high in et al7 reported no significant changes in TG, LDL, or
71 (22%) patients. At 3-month follow-up, mean HDL levels measured at baseline or during treatment
(SD) HDL levels were 48 (12) mg/dL, with low lev- with isotretinoin.
els in 63 (19.6%) patients, normal in 154 (47.8%) Some studies suggest that routine laboratory tests
patients, and high in 41 (12.7%) patients. At are needed when treating patients with isotretinoin
6-month follow-up, mean (SD) HDL levels were due to severe alterations in serum liver transaminase
47.6 (12) mg/dL, with low levels in 48 (14.9%) and lipid levels, while other studies conclude that
patients, normal in 117 (36.3%) patients, and high in the effects are minimal and laboratory tests are not
33 (10.2%) patients. Overall, statistically significant needed. In the current study, we found that there

WWW.CUTIS.COM VOLUME 94, NOVEMBER 2014 237

Copyright Cutis 2014. No part of this publication may be reproduced, stored, or transmitted without the prior written permission of the Publisher.
 Effects of Oral Isotretinoin

were statistically significant increases in TG and

LDL levels in patients who underwent treatment
with isotretinoin. We also found statistically sig-
nificant decreases in HDL levels. In our study, liver
enzymes were less affected than lipids in patients
who underwent treatment with isotretinoin. There
were statistically significant increases in AST levels,
but the clinical classification was not affected. There
also were increases in ALT levels, but the changes
were not statistically significant.
Overall, we advise dermatologists that isotreti-
noin can be administered with minimal concern
regarding changes in serum transaminase and lipid
levels; however, although severe laboratory altera-
tions were not noted in our study, we advise physi-
cians to use caution when administering isotretinoin
in patients with a history of abnormal findings.

y
op
References
1. Vieira AS, Beijamini V, Melchiors AC. The effect of
isotretinoin on triglyceride and liver aminotransferases.
An Bras Dermatol. 2012;87:382-387.
2. Brito MFM, Pessoa IS, Galindo JCS, et al. Evaluation
tc
of clinical adverse effects and laboratory alterations in
no
patients with acne vulgaris treated with oral isotretinoin [in
English, Portuguese]. An Bras Dermatol. 2010;85:331-337.
3. Expert Panel on Detection, Evaluation, and Treatment
o

of High Blood Cholesterol in Adults. Executive summary

D

of the third report of the National Cholesterol Education

Program (NCEP) Expert Panel on Detection, Evaluation,
and Treatment of High Blood Cholesterol In Adults (Adult
S

Treatment Panel III). JAMA. 2001;285:2486-2497.

4. Zane LT, Leyden WA, Marqueling AL, et al. A popula-
TI

tion-based analysis of laboratory abnormalities during

U

isotretinoin therapy for acne vulgaris. Arch Dermatol.

2006;142:1016-1022.
C

5. Bershad S, Rubinstein A, Paterniti JR, et al. Changes in

plasma lipids and lipoproteins during isotretinoin therapy
for acne. N Engl J Med. 1985;313:981-985.
6. Alcalay J, Landau M, Zucker A. Analysis of laboratory data
in acne patients treated with isotretinoin: is there really
a need to perform routine laboratory tests? J Dermatolog
Treat. 2001;12:9-12.
7. Baxter KF, Ling TC, Barth JH, et al. Retrospective survey of
serum lipids in patients receiving more than three courses
of isotretinoin. J Dermatolog Treat. 2003;14:216-218.

238 CUTIS® WWW.CUTIS.COM

Copyright Cutis 2014. No part of this publication may be reproduced, stored, or transmitted without the prior written permission of the Publisher.