Skip to main content
NCBI home page
Medicina logoLink to Medicina
. 2020 Oct 1;56(10):513. doi: 10.3390/medicina56100513

Hypnotics and Risk of Cancer: A Meta-Analysis of Observational Studies

Tzu-Rong Peng 1, Li-Jou Yang 1, Ta-Wei Wu 1,2, You-Chen Chao 3,4,*
PMCID: PMC7601941  PMID: 33019585

Abstract

Background and objectives: The association between hypnotic drugs and risk of cancer remains controversial. Therefore, we performed a meta-analysis to investigate this association. Materials and Methods: Pubmed and Embase were searched systematically to identify publications up to April 2020. The Newcastle-Ottawa scale for observational studies was used to assess the quality of studies. All included studies were evaluated by two reviewers independently; any discrepancies were resolved through discussion. Results: Twenty-eight studies including 22 case-control studies and 6 cohort studies with 340,614 hypnotics users and 1,828,057 non-users were included in the final analyses. Hypnotics (benzodiazepines and Z-drugs) use was significantly associated with an increased risk of cancer (odds ratio [OR] or relative risk [RR] 1.17; 95% confidence interval 1.09–1.26) in a random-effects meta-analysis of all studies. Subgroup meta-analysis by anxiolytics/sedatives effect (anxiolytics benzodiazepines vs. sedatives group (include sedatives benzodiazepines and Z-drugs)) revealed that a significant association in sedatives group (pooled OR/RR 1.26, 95% CI, 1.10–1.45), whereas no significant relationship was observed in anxiolytics benzodiazepines (pooled OR/RR 1.09, 95% CI, 0.95–1.26). Moreover, a significant dose–response relationship was observed between the use of hypnotics and the risk of cancer. Conclusions: This meta-analysis revealed association between use of hypnotics drugs and risk of cancer. However, the use of lower dose hypnotics and shorter duration exposed to hypnotics seemed to be not associated with an increased risk of cancer. Moreover, the use of anxiolytics effect benzodiazepines seemed to be lower risk than sedatives benzodiazepines. A high heterogeneity was observed among identified studies, and results were inconsistent in some subgroups. Randomized control trials are needed to confirm the findings in the future.

Keywords: hypnotic drugs, cancer, benzodiazepines

1. Introduction

Hypnotics (benzodiazepines and Z-drugs) are medications used to improve sleeping quality and to reduce wakefulness [1]. Benzodiazepines and Z-drugs (zolpidem and zopiclone) are common medications prescribed for sleep disorder. Moreover, benzodiazepines are prescribed in treating diseases such as seizures, anxiety, insomnia, and depression. Benzodiazepines derivatives have been distinguished into anxiolytics (i.e., diazepam, oxazepam, bromazepam, alprazepam, fludiazepam) and sedatives (i.e., flurazepam, flunitrazepam, estazolam, triazolam, temazepam, midazolam) by their effect. Z-drugs are non-benzodiazepines sedative-hypnotic medications commonly used to treat insomnia. The prevalence of benzodiazepine uses ranges from 10% to 43% worldwide among the aged population [2].

Meanwhile, most common adverse effects associated with hypnotics are residual daytime sedation, drowsiness, cognitive impairment, motor incoordination, dependence [3,4,5,6], and even tremor, which imply the possible neurotoxicity in long-term or high dose use of hypnotics. Several previous laboratory or animal studies had demonstrated that uses of benzodiazepine drugs or Z-drugs are risk factors for cancer [7].

Cancer is the second leading cause of death globally and leads to economic burden for health systems or costs for patients. Previous studies have assessed the relation between benzodiazepines or Z-drugs use and cancer risk. However, the results are still controversial. In addition, the quantitative meta-analysis has separately examined the benzodiazepines or Z-drugs and cancer risk. Until now, no published quantitative meta-analysis discussed benzodiazepines and Z-drugs long-term use conjunction with cancers. Recently, several observational studies with a larger population indicated conflicting inconsistent results. Combining the results of these studies in a meta-analysis may strengthen their statistical power. In this study, we updated the hypnotic drug correlation between use and the risk of cancer by using a meta-analysis of observational studies both case-control studies and prospective cohort studies.

2. Methods

2.1. Literature Search Strategy

PubMed and EMBASE were searched, by using selected keywords linked with benzodiazepine and non-benzodiazepine hypnotics and the risk of cancer up to April 2020. Keywords were as follows: benzodiazepine or zolpidem or zopiclone or diazepam or alprazolam or clonazepam or temazepam or oxazepam and cancer or tumor or carcinoma or neoplasm and case-control or cohort. Moreover, we reviewed the bibliographies of relevant articles to locate additional studies. The search was limited to human patients and articles in English. The detailed information on the search strategy for eligible studies is given in the flowchart provided by Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

2.2. Selection Relevant Studies and Criteria

The studies we included meet all the following criteria: (1) case-control or cohort study, (2) investigated the associations between the use of benzodiazepines or Z-drugs and the risk of cancer, (3) outcome measures with adjusted odds ratios (OR) or relative risks (RR) and 95% confidence intervals (CI). If data were duplicated or shared in more than one study, the longest-term follow-up studies were included in the analysis. We excluded non-published studies.

2.3. Data Extraction

We used a standardized data extraction form that included study year, study location, study population, participant characteristics, cancer types and crude or adjusted effect sizes and their 95% confidence intervals (CIs).

2.4. Risk of Bias of Included Studies

Newcastle-Ottawa Scale (NOS) was used to assess the methodological quality of included case-controls and cohort studies [8]. The quality score can range from 0 to 9. A study is classified in each domain as at high or low risk according to prespecified criteria (see Table 1). All included studies were evaluated by two pharmacists (T.W.W.) and (T.R.P.) independently; any discrepancies were resolved through discussion.

Table 1.

Methodological quality of the studies included in the final analysis (n = 28).

Case-Control
Studies (n = 21) 		Selection Comparability Control for
Important Factor
or Additional Factor 		Exposure Total
Adequate Definition
of Cases 		Representativeness
of Cases 		Selection
of Controls 		Definition
of Controls 		Ascertainment
of Exposure 		Same Method
of Ascertainment
for Participants 		Nonresponse
Rate
Kaufman (1982) 1 1 0 0 2 1 1 0 6
Kleinerman (1984) 1 1 1 0 2 1 1 0 7
Kaufman (1990) 1 1 1 0 2 1 1 0 7
Harlow (1995) 1 1 1 0 2 1 1 0 7
Rosenberg (1995) 1 1 0 1 2 1 1 0 7
Hardell (1996) 1 1 0 0 1 1 1 0 5
Westerdahl (1996) 1 1 1 0 2 1 1 0 7
Friedman (1998) 1 1 1 0 2 1 1 0 7
Coogan (2000) 1 1 0 0 2 1 1 0 6
Lagergen (2000) 1 1 1 0 2 1 1 0 7
Dublin (2002) 1 1 1 0 2 1 1 0 7
Pogoda (2004) 1 1 1 0 1 1 1 0 6
Halapy (2006) 1 1 1 0 2 1 1 0 7
Landgren (2006) 1 1 1 0 1 1 1 0 6
Fortuny (2007) 1 1 1 0 2 1 1 1 8
Pottegard (2012) 1 1 1 1 2 1 1 0 8
Iqbal (2014) 1 1 1 1 2 1 1 0 8
Hung (2016) 1 1 1 0 2 1 1 0 7
Lai (2017) 1 1 1 0 2 1 1 0 7
Thygesen (2017) 1 1 1 0 2 1 1 0 7
Lai (2019) 1 1 1 0 2 1 1 0 7
Cohort Studies
(n = 7) 		Selection Comparability Control for
Important
Factor
or Additional
Factor 		Outcome Total
Representativeness
of the Exposed
Cohort 		Selection of the Non-Exposed Cohort Ascertainment
of Exposure 		Outcome
of Interest
Was Not
Present
at Start of Study 		Assessment
of Outcome 		Follow-up
Long Enough
for Outcomes
to Occur 		Adequacy
of Follow-up
of Cohorts
Kripke (2012) 1 1 1 1 2 1 1 0 8
Kao_B (2012) 1 1 1 1 1 1 0 0 6
Jaussent (2013) 1 1 1 1 2 1 1 0 8
Harnod_B (2014) 1 1 1 1 1 1 1 0 7
Kao_Z (2012) 1 1 1 1 1 1 1 0 7
Harnod (2015) 1 1 1 1 1 1 1 0 7
Lin (2016) 1 1 1 1 1 1 1 0 7
Open in a new tab

2.5. Statistical Analyses

We conducted meta-analyses on the association of hypnotics use and the risk of cancer by observational studies. The random-effects model (DerSimonian–Laird method) was used to calculate the pooled OR or RR [9]. The Cochran Q test and I2 statistics were used to assess statistical heterogeneity and inconsistency. Statistical significance was set at p < 0.10 for Cochrane Q tests. Heterogeneity was considered low, moderate, or high, if the I2 values was < 25%, 25–50%, and > 50%, respectively. Results were considered as statistically significant when the p value was less than 0.05. Publication bias was examined by using funnel plots, and Egger’s and Begg’s test was used to analyze the publication bias in our studies. A p-value > 0.05 based on the Egger’s and Begg’s test indicated the absence of publication bias. Statistical analysis was performed according to the Cochrane Handbook for Statistical Review of Interventions (version 5.4) [10]. The meta-analysis was performed by using RevMan software (The Cochrane Collaboration, Oxford, UK) and STATA version 15.0 (StataCorp, College Station, TX, USA).

3. Results

3.1. Study Characteristics

A total of 658 records were screened, and 113 full-text articles were assessed for eligibility. Twenty-eight articles were selected for qualitative review, including 22 case-control studies and 6 cohort studies (Figure 1). The characteristics of the 28 included studies are summarized in Table 2. The studies involved 2,168,671 participants (340,614 hypnotics users and 1,828,057 non-users).

Figure 1.

Figure 1

Open in a new tab

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2009 flow diagram.

Table 2.

Characteristics of included studies.

Author (year) Study type Country Years
Enrolled		 Population
(Hypnotic
/Control)		 Cancer Type
 Definition of Hypnotic
use		 OR/RR
(95% CI)
Kaufman (1982) [11] Case-control Canada, United States and Israel 1976–1980 1236/728 Breast Diazepam ≥ 6 month vs. never-use 0.9 (0.5–1.6)
Kleinerman (1984) [12] Case-control United States 1973–1977 1075/1146 Breast Diazepam ≥ 6 month vs. never-use 0.81 (0.6–1.1)
Kaufman (1990) [13] Case-control United States 1981–1987 3078/1931 Breast Diazepam ≥ 6 month vs. never-use 1.0 (0.6–1.7)
Kaufman_1 (1990) [13] Case-control Canada 1982–1986 607/1214 Breast Diazepam ≥ 6 month vs. never-use 0.8 (0.5–1.3)
Harlow (1995) [14] Case-control United States 1978–1987 450/454 Ovarian Benzodiazepine vs. never-use 1.8 (1.0–3.1)
Rosenberg (1995) [15] Case-control United States 1977–1991 382/5695 Non-Hodgkin’s lymphoma Benzodiazepine ≥ 1 month vs. never- use 2.1 (1.4–3.3)
Hardell (1996) [16] Case-control Sweden 1984–1986 329/658 Colon Benzodiazepine vs. never-use 1.7 (0.9-3.3)
Westerdahl (1996) [17] Case-control Sweden 1988–1990 400/640 Malignant melanoma Benzodiazepine vs. never-use 1.8 (0.7–4.4)
Friedman (1998) [18] Case-control United States 1991–1994 1993/2410 Colon Diazepam ≥ 12 month vs. never-use 1.2 (0.8–1.8)
Coogan (2000) [19] Case-control United States 1976–1998 748/2992 Ovarian Benzodiazepine < 12 month vs. never-use 1.4 (1.0–2.1)
Lagergen (2000) [20] Case-control Sweden 1995–1997 189/820 Esophageal Benzodiazepine vs. never-use 1.5 (0.7–2.9)
Dublin (2002) [21] Case-control United States 1981–1997 314/790 Ovarian Benzodiazepine < 6 month vs. never-use 0.70 (0.47–1.0)
Pogoda (2004) [22] Case-control United States 1987–1994 412/412 Acute myeloid leukemia Benzodiazepine ≥ 6 month vs. never-use 1.5 (0.6–3.7)
Halapy (2006) [23] Case-control Canada 1996–1998 3133/3062 Breast Benzodiazepine vs. never-use 1.06 (0.88–1.27)
Landgren (2006) [24] Case-control United States 1997–2002 179/691 Multiple myeloma Benzodiazepine ≥ 6 month vs. never-use 0.9 (0.3–2.6)
Fortuny (2007) [25] Case-control United States 1980–2002 114/3996 Esophageal Benzodiazepine vs. never-use 1.7 (0.9–3.1)
Kripke (2012) [26] Prospective cohort United States 2002–2007 2076 cases
among 25,750		 All cancers Any hypnotic > 132 pill/year vs. non-users 1.35 (1.18–1.55)
Kao_B (2012) [27] Prospective cohort Taiwan 1996–2000 3520 cases
among
119,239		 All cancers Benzodiazepine ≥ 2 month vs. non-users 1.19 (1.08–1.32)
Kao_Z (2012) [28] Prospective cohort Taiwan 1998–2000 1047/2924 All cancers Zolpidem vs. never-use 1.68 (1.55–1.82)
Pottegard (2012) [29] Case-control Denmark 2002–2009 149360/1194729 All cancers All benzodiazepine any related drugs (cumulative amount ≥ 500 defined daily dose) vs. never use 1.09 (1.04–1.14)
Jaussent (2013) [30] Prospective cohort France 1999–2011 1454 cases
among 6696		 All cancers Hypnotic vs. never-use 0.96 (0.74–1.23)
Harnod_B (2014) [31] Prospective cohort Taiwan 2000–2009 274 cases
among
62,050		 Brain cancer Benzodiazepine ≥ 2 month vs. never-use 3.15 (2.37–4.20)
Harnod_Z (2015) [32] Prospective cohort Taiwan 2000–2009 37810/37810 Brain cancer Zolpidem ≥ 520 mg/year vs. never-use 1.85 (1.21–2.82)
Iqbal (2014) [33] Case-control Taiwan 1998–2009 42500/255000 All cancers Benzodiazepine ≥ 2 month vs. never-use 1.21 (1.18–1.24)
Hung (2016) [34] Case-control Taiwan 2006–2011 1454/1448 Hepatocellular carcinoma Clonazepam vs. never-use 0.93 (0.76–1.13)
Lin (2016) [35] Prospective cohort Taiwan 2002–2004 1728 cases
among 6924		 All cancers Zolpidem vs. never-use 1.75 (1.02–3.0)
Lai (2017) [36] Case-control Taiwan 2011–2012 77986/77986 Hepatocellular carcinoma Benzodiazepine vs. never-use 1.5 (1.45–2.44)
Thygesen (2017) [37] Case-control Danish 2002–2009 1854/4950 All cancers Benzodiazepine > 500 DDD (1–5 years) vs. never-use 1.09 (1.00–1.19)
Lai (2019) [38] Case-control Taiwan 2000–2013 4912/4912 Colorectal Zolpidem vs. never-use 1.05 (0.95–1.15)
Open in a new tab

Abbreviation: OR, odds ratio; RR, relative risk; CI, confidence interval; DDD, defined daily dose.

3.2. Quality Assessment

We performed the methodological quality of studies based on NOS scales. The results of the methodological quality of studies are summarized in Table 2. The NOS range from 5–8; average NOS score was 6.9 and 7.1 for case-control studies and cohort studies, respectively; In case-control studies, 3 high-quality studies are included [25,29,33] (score of 8); in cohort studies, 6 high-quality studies are included [26,28,30,31,32,35] (score of 7).

3.3. Meta-Analysis and Subgroup Analysis

Overall, the risk of cancer was greater in hypnotics user than nonusers in the random-effects meta-analysis of all 27 studies (pooled OR/RR, 1.17; 95% CI, 1.09–1.26, p < 0.001, I2 = 79%) (Figure 2).

Figure 2.

Figure 2

Open in a new tab

Forest plot of benzodiazepines/Z-drugs use and the risk of cancer.

Subgroup analyses were carried out by research methods design (cohort or case-control study) and hypnotics categories (benzodiazepines, Z-drugs). In the subgroup analyses aligned with study design, hypnotics showed significant positive correlation with risk of cancer, both in case-control and cohort studies subgroup (pooled OR/RR was 1.10; 95% CI, 1.03–1.18 for 22 case-control studies and 1.50; 95% CI, 1.11–2.02 for 5 cohort studies) (Figure 3). In the subgroup analyses of hypnotics categories (benzodiazepines vs. Z-drugs), 24 studies are included in benzodiazepines group, the pooled OR/RR was 1.19 (95% CI, 1.10–1.29), and 8 studies are included in Z-drugs group and the pooled OR/RR was 1.24 (95% CI, 1.08–1.42) (Figure 4). Z-drugs group showed higher OR/RR than benzodiazepines group.

Figure 3.

Figure 3

Open in a new tab

Forest plot of benzodiazepine/Z-drugs use and the risk of cancer by type of study design.

Figure 4.

Figure 4

Open in a new tab

Forest plot of use of benzodiazepines vs. Z-drugs and the risk of cancer.

3.4. Subgroup Meta-Analyses by Different Factors

The results of subgroup meta-analyses by different factors were showed in Table 3. The use of hypnotics was associated with an increased risk of cancer in the subgroup meta-analyses by duration of hypnotics use, cumulative yearly dose, and sedatives benzodiazepines and Z-drugs. Subgroup meta-analysis by gender showed that a significant association in both gender with 19 studies (pooled OR/RR 1.22, 95% CI, 1.13–1.32; I2 = 42%), whereas no significant relationship was observed in female with nine studies (pooled OR/RR 1.01, 95% CI, 0.89–1.14; I2 = 83%). However, there was no significant relationship observed in four studies with only elderly subjects (pooled OR/RR 1.16, 95% CI, 0.92–1.47). Subgroup meta-analysis by anxiolytics/sedatives effect (anxiolytics benzodiazepines vs. sedatives group (include sedatives benzodiazepines and Z-drugs)) revealed a significant association in sedatives group (pooled OR/RR 1.26, 95% CI, 1.10–1.45), whereas no significant relationship was observed in anxiolytics benzodiazepines (pooled OR/RR 1.09, 95% CI, 0.95–1.26). Regarding the type of hypnotics, short-acting hypnotics (midazolam, oxazepam, alprazolam, triazolam, and zolpidem) showed a significantly increased risk. Hypnotics use increased the risk of brain cancer, esophagus cancer, liver cancer, lung cancer, stomach cancer, pancreatic cancer, colon cancer, renal cancer, and prostate cancer. However, no significant association was observed in malignant melanoma, breast cancer, and ovarian cancer. When compared with nonusers of hypnotics, the pooled OR/RR for the risk of cancer was 1.03 (95% CI, 1.01–1.05) in a low dose, 1.30 (95% CI, 0.97–1.75) in a medium dose, and 2.03 (95% CI, 1.19–3.46) in a high dose (Table 3).

Table 3.

Benzodiazepines/Z-drugs use and the risk of cancer in the subgroup meta-analysis by different factors.

Factors Study Number Summary OR or RR (95% CI) Heterogeneity I2 (%) Random/Fixed Effects
All 27 1.17 (1.09–1.26) 79% Random
Region
America 14 1.15 (0.95–1.36) 62% Random
Europe 6 1.09 (1.05–1.14) 0% Random
Asia 7 1.24 (1.09–1.42) 91% Random
Type of cancer
Brain cancer 5 1.93 (1.29–2.88) 82% Random
Malignant melanoma 4 1.01 (0.78–1.31) 0% Fixed
Esophagus cancer 6 1.56 (1.32–1.84) 0% Fixed
Breast cancer 10 1.08 (0.96–1.22) 61% Random
Liver cancer 7 1.38 (1.17–1.63) 89% Random
Lung cancer 5 1.24 (1.04–1.48) 80% Random
Stomach cancer 3 1.18 (1.05–1.34) 4% Fixed
Pancreatic cancer 2 1.38 (1.20–1.58) 0% Fixed
Colon cancer 7 1.11 (1.01–1.23) 59% Random
Ovarian cancer 7 1.07 (0.86–1.33) 50% Random
Renal cancer 4 1.51 (1.18–1.94) 60% Random
Prostate cancer 4 1.29 (1.07–1.55) 70% Random
Gender
Female 9 1.01 (0.89–1.14) 42% Fixed
Male and Female 19 1.22 (1.13–1.32) 83% Random
Elderly ≥ 65 4 1.16 (0.92–1.47) 84% Random
Anxiolytics/ Sedatives
Anxiolytics benzodiazepines 9 1.09 (0.95–1.26) 30% Random
Sedatives benzodiazepines and Z-drugs 10 1.26 (1.10–1.45) 93% Random
Duration of hypnotics use
<6 months 12 1.03 (1.02–1.04) 35% Fixed
≥6 months 13 1.05 (1.02–1.08) 0% Fixed
≥5 years 9 1.11 (1.02–1.21) 0% Fixed
Cumulative yearly dose
Lower 5 1.03 (1.01–1.05) 10% Fixed
Moderate 6 1.30 (0.97–1.75) 95% Random
Highest 6 2.03 (1.19–3.46) 97% Random
Type of hypnotics
Long-acting (Diazepam) 8 0.97 (0.93–1.01) 37% Fixed
Intermediate-acting 4 1.21 (0.93–1.57) 88% Random
Short-acting 9 1.29 (1.12–1.48) 92% Random
Methodological quality
High quality 19 1.14 (1.04–1.25) 92% Random
Low quality 10 1.59 (1.27–1.98) 85% Random
Open in a new tab

3.5. Publication Bias

A visual inspection of the funnel plot of OR/RR from these studies revealed asymmetry (Figure 5). However, both the Egger’s and Begg’s test suggested no statistical evidence of publication bias, with p value of 0.541 and 0.420, respectively.

Figure 5.

Figure 5

Open in a new tab

Publication bias funnel plots for use of benzodiazepines/Z-drugs and the risk of cancer.

4. Discussion

In our updated meta-analysis studies, the use of hypnotics was associated with an increased risk of cancer. Subgroup meta-analyses by different factors also showed similar results. However, this meta-analysis revealed that the cancer risk is related to dose–response, sedatives benzodiazepines and Z-drugs (pooled OR/RR 1.26, 95% CI, 1.10–1.45), and duration of hypnotics use (long term use: pooled OR/RR 1.11, 95% CI, 1.02–1.21). In recent years, several literatures have reported a tentative link between benzodiazepines and/or Z-drugs exposure with adverse outcomes such as respiratory disease exacerbation, infections, inflammation, dementia, pancreatitis, and cancer [39]. These kinds of adverse outcome, especially hypnotics, may relate to inflammation, infection, or cancer patients suffering from psychiatric and leading to more benzodiazepines exposure might reflect sharp of cancer diagnosed.

According to a review by Brambilla et al., the fact that the mechanisms of benzodiazepines- and Z-drugs-induced tumorigenesis remains tentative and unclear [40]. A study has reported evidence that use of hypnotics may lead to decline in immune function. Several animal studies have revealed that benzodiazepines disrupted the processes of phagocytes spreading and macrophages oxidative bursting [41,42]. These may be reduced release of the proinflammatory cytokines interleukin-6 and interleukin-13 in blood cells because of the activation of their benzodiazepine receptors [43]. Hypnotics use showed the strongest association with the risk of brain cancer in this meta-analysis (Table 3). These findings were consistent with results of Kim et al. (2017) and Zhang et al. (2017) [2,44]. The possible mechanism is that hypnotics enhance the neurotransmitter of gamma-aminobutyric acid (GABA) by interacting with the chlorine ion channel that binds to GABA receptors. The gamma-aminobutyric acid has an inhibitory neurotransmitter effect but also can regulate cell proliferation and differentiation of brain and peripheral at various stages and may participate benign tumor growth [45,46]. However, these potential mechanisms are still unproven.

Currently, there is still a lack of conclusive experimental data, but alarm signals for cancer risk have been raised by researchers for hypnotic drugs based on observational study findings [26,27,28,33]. A previous meta-analysis of 22 observational studies (18 case-control and 4 cohort studies) concluded there is an overall estimate of 19% increased cancer risk, with a significant dose–response trend, among benzodiazepines users over non-users [47]. A meta-analysis performed by Kim et al. (2017) included 6 observational epidemiological studies (3 case-control and 3 cohort studies) [44]. In a meta-analysis, compared with non-use of hypnotics, the OR for overall hypnotics (zopiclone or zolpidem) use was 1.29 for various cancers (95% confidence interval, 1.08–1.53). Our meta-analysis reported an overall estimate of 17% increased cancer risk with benzodiazepines- and Z-drugs use. Z-drugs group showed higher OR/RR (1.24, 95% CI, 1.08–1.42) than benzodiazepines group (1.19, 95% CI, 1.10–1.29). Our meta-analysis showed that hypnotics use was associated with the increased risk of cancer by duration of hypnotics, highest cumulative dose, and short acting hypnotics. According to the Bradford Hill criteria, the biological gradient (dose–response) is one of the important criteria confirming a causal relationship [48]. As shown, the dose of hypnotic is related to cancer risk, but it can also increase the causal relationship. The elderly is usually defined as individuals aged 65 years and older [49]. Insomnia is problematic for older adults. Two-fold increase was found in the intake of hypnotics among the elderly [50]. Current available hypnotic drugs all have significant risks for the elderly, such as increased risk for falls and cognitive function decline. However, we found that hypnotics use among the elderly showed a trend towards increased risk of cancer but not significantly (OR/RR, 1.16; 95% CI, 0.92–1.47; I2 = 84%). It may be associated with the small sample size (4 trials). Iqbal et al.’s (2014) study uses the Taiwanese National Health Insurance system to gather information about benzodiazepines use and cancer risk. They found that clonazepam, lorazepam, alprazolam, bromazepam, zolpidem, and zopiclone have a high risk of cancer [33]. However, we found that sedatives benzodiazepines and Z-drugs have higher cancer risk (pooled OR/RR 1.26, 95% CI, 1.10–1.45). Moreover, Z-drugs (zolpidem and zopiclone) have higher cancer risk than benzodiazepines (24% vs. 19%).

In addition, the most common risk factors for cancer including aging, smoking, alcohol consumption, family history, and exposure to chemicals or other substances [51]. The studies included in this meta-analysis, all of them, adjusted age as a confounding factor; fourteen studies adjusted alcohol drinking as a confounding factors; nine studies adjusted tobacco smoking as a confounding factor; nine studies adjusted family cancer history as a confounding factor, and only six studies adjusted medical use as a confounding factor. Thus, our studies are not excluding some important confounding factors such as alcohol drinking, tobacco smoking, family cancer history, and medical use that association between the hypnotics use and cancer risk. Nevertheless, there are still many factors that contribute to the risk factors of cancer, which have not been corrected. Therefore, the results should not be over-interpreted.

This study has some limitations. First, a portion of the included studies in our analysis did not adjust tobacco smoking, alcohol drinking, family cancer history, and medical use factors, which are known as important factors related to cancer. Second, this study only included cohort studies and case-control studies because there are no randomized controlled trials published on this topic. However, cohort studies and case-control studies have a lower level of evidence than randomized controlled trials.

5. Conclusions

The present meta-analysis found that hypnotics use was associated with an increased risk of cancer. However, the use of lower dose hypnotics and shorter duration exposed to hypnotics seemed to be not associated with an increased risk of cancer. Moreover, the use of anxiolytic effect benzodiazepines seemed to be lower risk than sedative benzodiazepines. Further large randomized controlled trials providing a higher level of evidence should be conducted to confirm our findings.

Author Contributions

Conceptualization, T.-R.P. and T.-W.W.; methodology, T.-R.P.; software, T.-R.P.; validation, T.-R.P., L.-J.Y., and T.-W.W.; formal analysis, T.-R.P. and L.-J.Y.; investigation, T.-W.W.; data curation, L.-J.Y.; writing—original draft preparation, T.-R.P.; writing—review and editing, T.-W.W. and Y.-C.C. All authors have read and agreed to the published version of the manuscript.

Funding

This study was supported by grants from the Taipei Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation (TCRD-TPE-109-18).

Conflicts of Interest

The authors declare no conflict of interest.

References

  • 1.Deoras K., Moul D. Encyclopedia of the Neurological Sciences. 2nd ed. Academic Press; Cambridge, MA, USA: 2014. Hypnotics; pp. 646–649. [Google Scholar]
  • 2.Zhang T., Yang X., Zhou J., Liu P., Wang H., Li A., Zhou Y. Benzodiazepine drug use and cancer risk: A dose-response meta analysis of prospective cohort studies. Oncotarget. 2017;8:102381–102391. doi: 10.18632/oncotarget.22057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.National Institutes of Health National Institutes of Health State of the Science Conference statement on Manifestations and Management of Chronic Insomnia in Adults. Sleep. 2005;28:1049. doi: 10.1093/sleep/28.9.1049. [DOI] [PubMed] [Google Scholar]
  • 4.Buscemi N., Vandermeer B., Friesen C., Bialy L., Tubman M., Ospina M., Klassen T.P., Witmans M. The efficacy and safety of drug treatments for chronic insomnia in adults: A meta-analysis of RCTs. J. Gen Intern Med. 2007;22:1335. doi: 10.1007/s11606-007-0251-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Holbrook A.M., Crowther R., Lotter A., Cheng C., King D. Meta-analysis of benzodiazepine use in the treatment of insomnia. CMAJ. 2000;162:225. [PMC free article] [PubMed] [Google Scholar]
  • 6.Nowell P.D., Mazumdar S., Buysse D.J., Dew M.A., Reynolds C.F., Kupfer D.J. Benzodiazepines and zolpidem for chronic insomnia: A meta-analysis of treatment efficacy. JAMA. 1997;278:2170. doi: 10.1001/jama.1997.03550240060035. [DOI] [PubMed] [Google Scholar]
  • 7.Kunert-Radek J., Stepien H., Pawlikowski M. Inhibition of rat pituitary tumor cell proliferation by benzodiazepines in vitro. Neuroendocrinology. 1994;59:92–96. doi: 10.1159/000126644. [DOI] [PubMed] [Google Scholar]
  • 8.Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur. J. Epidemiol. 2010;25:603–605. doi: 10.1007/s10654-010-9491-z. [DOI] [PubMed] [Google Scholar]
  • 9.DerSimonian R., Laird N. Meta-analysis in clinical trials. Control. Clin. Trials. 1986;7:177–188. doi: 10.1016/0197-2456(86)90046-2. [DOI] [PubMed] [Google Scholar]
  • 10.Higgins J., Deeks J.J., Altman D.G. Cochrane Book Series. Wiley & Sons; London, UK: 2008. Special Topics in Statistics. Cochrane Handbook for Systematic Reviews of Interventions; pp. 48–529. [Google Scholar]
  • 11.Kaufman D.W., Shapiro S., Slone D., Rosenberg L., Helmrich S.P., Miettinen O.S., Stolley P.D., Levy M., Schottenfeld D. Diazepam and the risk of breast cancer. Lancet. 1982;6:537–539. doi: 10.1016/S0140-6736(82)92046-3. [DOI] [PubMed] [Google Scholar]
  • 12.Kleinerman R.A., Brinton L.A., Hoover R., Fraumeni J.F. Diazepam use and progression of breast cancer. Cancer Res. 1984;44:1223–1225. [PubMed] [Google Scholar]
  • 13.Kaufman D.W., Werler M.M., Palmer J.R., Rosenberg L., Stolley P.D., Shapiro S., Warshauer M.E., Clarke E.A., Miller D.R. Diazepam use in relation to breast cancer: Results from two case-control studies. Am. J. Epidemiol. 1990;131:483–490. doi: 10.1093/oxfordjournals.aje.a115522. [DOI] [PubMed] [Google Scholar]
  • 14.Harlow B.L., Cramer D.W. Self-reported use of antidepressants or benzodiazepine tranquilizers and risk of epithelial ovarian cancer: Evidence from two combined case-control studies (Massachusetts, United States) Cancer Causes Control. 1995;6:130–134. doi: 10.1007/BF00052773. [DOI] [PubMed] [Google Scholar]
  • 15.Rosenberg L., Palmer J.R., Zauber A.G., Warshauer M.A., Strom B.L., Hariap S., Shapiro S. Relation of benzodiazepine use to the risk of selected cancers: Breast, large bowel, malignant melanoma, lung, endometrium, ovary, non-Hodgkin’s lymphoma, testis, Hodgkin’s disease, thyroid, and liver. Am. J. Epidemiol. 1995;141:1153–1160. doi: 10.1093/oxfordjournals.aje.a117388. [DOI] [PubMed] [Google Scholar]
  • 16.Hardell L., Fredrikson M., Axelson O. Case-control study on colon cancer regarding previous diseases and drug intake. Int. J. Oncol. 1996;8:439–444. doi: 10.3892/ijo.8.3.439. [DOI] [PubMed] [Google Scholar]
  • 17.Westerdahl J., Olsson H., Måskback A., Ingvar C., Jonsson N. Risk of malignant melanoma in relation to drug intake, alcohol, smoking and hormonal factors. Br. J. Cancer. 1996;73:1126–1131. doi: 10.1038/bjc.1996.216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Friedman G.D., Coates A.O., Potter J.D., Slattery M.L. Drugs and colon cancer. Pharmacoepidemiol. Drug Saf. 1998;7:99–106. doi: 10.1002/(SICI)1099-1557(199803/04)7:2&#x0003c;99::AID-PDS320&#x0003e;3.0.CO;2-0. [DOI] [PubMed] [Google Scholar]
  • 19.Coogan P.F., Rosenberg L., Palmer J.R., Strom B.L., Stolley P.D., Zauber A.G., Shapiro S. Risk of ovarian cancer according to use of antidepressants, phenothiazines, and benzodiazepines (United States) Cancer Causes Control. 2000;11:839–845. doi: 10.1023/A:1008982417022. [DOI] [PubMed] [Google Scholar]
  • 20.Lagergren J., Bergstrom R., Adami H.O., Nyrén O. Association between medications that relax the lower esophageal sphincter and risk for esophageal adenocarcinoma. Ann. Intern. Med. 2000;133:165–175. doi: 10.7326/0003-4819-133-3-200008010-00007. [DOI] [PubMed] [Google Scholar]
  • 21.Dublin S., Rossing M.A., Heckbert S.R., Goff B.A., Weiss N.S. Risk of epithelial ovarian cancer in relation to use of antidepressants, benzodiazepines, and other centrally acting medications. Cancer Causes Control. 2002;13:35–45. doi: 10.1023/A:1013969611593. [DOI] [PubMed] [Google Scholar]
  • 22.Pogoda J.M., Katz J., McKean-Cowdin R., Nichols P.W., Ross R.K., Preston-Martin S. Prescription drug use and risk of acute myeloid leukemia by French-American-British subtype: Results from a Los Angeles County case-control study. Int. J. Cancer. 2005;114:634–638. doi: 10.1002/ijc.20788. [DOI] [PubMed] [Google Scholar]
  • 23.Halapy E., Kreiger N., Cotterchio M., Sloan M. Benzodiazepines and risk for breast cancer. Ann. Epidemiol. 2006;16:632–636. doi: 10.1016/j.annepidem.2005.11.004. [DOI] [PubMed] [Google Scholar]
  • 24.Landgren O., Zhang Y., Zahm S.H., Inskip P., Zheng T., Baris D. Risk of multiple myeloma following medication use and medical conditions: A case-control study in Connecticut women. Cancer Epidemiol. Biomark. Prev. 2006;15:2342–2347. doi: 10.1158/1055-9965.EPI-06-0097. [DOI] [PubMed] [Google Scholar]
  • 25.Fortuny J., Johnson C.C., Bohlke K., Chow W.-H., Hart G., Kucera G., Mujumdar U., Ownby D., Wells K., Engel L.S., et al. Use of anti-inflammatory drugs and lower esophageal sphincter-relaxing drugs and risk of esophageal and gastric cancers. Clin. Gastroenterol. Hepatol. 2007;5:1154–1159. doi: 10.1016/j.cgh.2007.05.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Kripke D.F., Langer R.D., Kline L.E. Hypnotics’ association with mortality or cancer: A matched cohort study. BMJ Open. 2012;2:e000850. doi: 10.1136/bmjopen-2012-000850. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Kao C.H., Sun L.M., Su K.P., Chang S.N., Sung F.C., Muo C.H., Liang J.A. Benzodiazepine use possibly increases cancer risk: A population-based retrospective cohort study in Taiwan. J. Clin. Psychiatr. 2012;73:e555–e560. doi: 10.4088/JCP.11m07333. [DOI] [PubMed] [Google Scholar]
  • 28.Kao C.H., Sun L.M., Liang J.A., Chang S.N., Sung F.C., Muo C.H. Relationship of zolpidem and cancer risk: A Taiwanese population-based cohort study. Mayo Clin. Proc. 2012;87:430–436. doi: 10.1016/j.mayocp.2012.02.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Pottegård A., Friis S., Andersen M., Hallas J. Use of benzodiazepines or benzodiazepine related drugs and the risk of cancer: A population-based case control study. Br. J. Clin. Pharmacol. 2013;75:1356–1364. doi: 10.1111/bcp.12001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Jaussent I., Ancelin M.L., Berr C., Pérès K., Scali J., Besset A., Ritchie K., Dauvilliers Y. Hypnotics and mortality in an elderly general population: A 12-year prospective study. BMC Med. 2013;11:212. doi: 10.1186/1741-7015-11-212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Harnod T., Lin C.L., Sung F.C., Kao C.H. An association between benzodiazepine use and occurrence of benign brain tumors. J. Neurol. Sci. 2014;336:8–12. doi: 10.1016/j.jns.2013.11.009. [DOI] [PubMed] [Google Scholar]
  • 32.Harnod T., Li Y.F., Lin C.L., Chang S.N., Sung F.C., Kao C.H. Higher-dose uses of zolpidem will increase the subsequent risk of developing benign brain tumors. J. Neuropsychiatr. Clin. Neurosci. 2015;27:e107–e111. doi: 10.1176/appi.neuropsych.14010006. [DOI] [PubMed] [Google Scholar]
  • 33.Iqbal U., Nguyen P.A., Syed-Abdul S., Yang H.-S., Huang C.-H., Jian W.-E., Hsu M.-I., Yen Y., Li Y.-C. Is longterm use of benzodiazepine a risk for cancer? Medicine. 2015;94:e483. doi: 10.1097/MD.0000000000000483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Hung D.Z., Lin C.L., Li Y.W. Association between antiepileptic drugs and hepatocellular carcinoma in patients with epilepsy: A population-based case-control study. Brain Behav. 2016;6:e00554. doi: 10.1002/brb3.554. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Lin S.C., Su Y.C., Huang Y.S., Lee C.C. Zolpidem increased cancer risk in patients with sleep disorder: A 3-year follow-up study. J. Med. Sci. 2016;36:68–74. [Google Scholar]
  • 36.Lai S.W., Lin C.L., Liao K.F. Zolpidem administration and risk of hepatocellular carcinoma: A case-control study in Taiwan. Front. Pharmacol. 2017;8:767. doi: 10.3389/fphar.2017.00767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Thygesen L.C., Pottegård A., Ersbøll A.K., Friis S., Stürmer T., Hallas J. External adjustment of unmeasured confounders in a case-control study of benzodiazepine use and cancer risk. Br. J. Clin. Pharmacol. 2017;83:2517–2527. doi: 10.1111/bcp.13342. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Lai S.W., Lin C.L., Liao K.F. Association between colorectal cancer and zolpidem use in a case-control study. Medicine. 2019;98:e18127. doi: 10.1097/MD.0000000000018127. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Brandt J., Leong C. Benzodiazepines and Z-drugs: An updated review of major adverse outcomes reported on in epidemiologic research. Drugs R D. 2017;17:493–507. doi: 10.1007/s40268-017-0207-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Brambilla G., Carrozzino R., Martelli A. Genotoxicity and carcinogenicity studies of benzodiazepines. Pharmacol. Res. 2007;56:443–458. doi: 10.1016/j.phrs.2007.08.006. [DOI] [PubMed] [Google Scholar]
  • 41.Massoco C., Palermo-Neto J. Effects of midazolam on equine innate immune response: A flow cytometric study. Vet. Immunol. Immunopathol. 2003;95:11–19. doi: 10.1016/S0165-2427(03)00097-7. [DOI] [PubMed] [Google Scholar]
  • 42.Massoco C.O., Palermo-Neto J. Diazepam effects of peritoneal macrophage activity and corticosterone serum levels in Balb/C mice. Life Sci. 1999;65:2157–2165. doi: 10.1016/S0024-3205(99)00481-6. [DOI] [PubMed] [Google Scholar]
  • 43.Torres S.R., Frode T.S., Nardi G.M., Vita N., Reeb R., Ferrara P., Farges R.C., Ribeiro-do-Valle R.M. Anti-inflammatory effects of peripheral benzodiazepine receptor ligands in two mouse models of inflammation. Eur. J. Pharmacol. 2000;408:199–211. doi: 10.1016/S0014-2999(00)00760-3. [DOI] [PubMed] [Google Scholar]
  • 44.Kim D.H., Kim H.B., Kim Y.H., Kim J.Y. Use of Hypnotics and Risk of Cancer: A Meta-Analysis of Observational Studies. Korean J. Fam. Med. 2018;39:211–218. doi: 10.4082/kjfm.17.0025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Enna S.J. Role of gamma-aminobutyric acid in anxiety. Psychopathology. 1984;17:15–24. doi: 10.1159/000284073. [DOI] [PubMed] [Google Scholar]
  • 46.Jezewska E., Scinska A., Kukwa W., Sobolewska A., Turzynska D., Samochowiec J., Bienkowski P. Gamma-aminobutyric acid concentrations in benign parotid tumours and unstimulated parotid saliva. J. Laryngol. Otol. 2011;125:492–496. doi: 10.1017/S0022215110002574. [DOI] [PubMed] [Google Scholar]
  • 47.Kim H.B., Myung S.K., Park Y.C., Park B. Use of benzodiazepine and risk of cancer: A meta-analysis of observational studies. Int. J. Cancer. 2017;140:513–525. doi: 10.1002/ijc.30443. [DOI] [PubMed] [Google Scholar]
  • 48.Hill A.B. The Environment and Disease: Association or Causation? Proc. R. Soc. Med. 1965;58:295–300. doi: 10.1177/003591576505800503. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Schutte-Rodin S., Broch L., Buysse D., Dorsey C., Sateya M. Clinical guideline for the evaluation and chronic management of chronic insomnia in adults. J. Clin. Sleep Med. 2008;4:487–504. doi: 10.5664/jcsm.27286. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Ohayon M.M., Caulet M., Priest R.G., Guilleminault C. Psychotropic medication consumption patterns in the UK general population. J. Clin. Epidemiol. 1998;51:273–283. doi: 10.1016/S0895-4356(97)00238-2. [DOI] [PubMed] [Google Scholar]
  • 51.White M.C., Holman D.M., Boehm J.E., Peipins L.A., Grossman M., Henley S.J. Age and cancer risk: A potentially modifiable relationship. Am. J. Prev. Med. 2014;46:S7–S15. doi: 10.1016/j.amepre.2013.10.029. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Medicina are provided here courtesy of Multidisciplinary Digital Publishing Institute (MDPI)

RESOURCES