The Burden of Hidradenitis Suppurativa Signs and Symptoms in Quality of Life: Systematic Review and Meta-Analysis

Hidradenitis suppurativa (HS) is a chronic, recurrent and debilitating inflammatory skin disease of the hair follicle that usually presents as painful, deep-seated inflamed lesions in the apocrine gland-bearing areas of the body. HS patients suffer from uncomfortable signs and symptoms, such as pain, pruritus, malodour and suppuration, which may impair patients’ quality of life (QoL). Although HS patients frequently experience these signs and symptoms, they are only occasionally assessed by clinicians and, unexpectedly, the scientific evidence available is limited and heterogeneous. The aim of this study is to summarize the evidence regarding the impact of HS signs and symptoms on QoL to serve as a basis for future research and help clinicians to consider them in the daily care of HS patients. A systematic review and meta-analysis were conducted following PRISMA Guidelines. The following search algorithm was used: (hidradenitis or “acne inversa”) and (pain or itch or odour or malodour or suppuration or oozing or drainage) and (“quality of life”). The literature search identified 836 references, 17 of them met the eligible criteria and were included for analysis, representing 4929 HS patients. Mean age of the participants was 36.28 years and there was a predominance of female sex among study participants. The BMI of the population was in the range of over-weight and about two out five patients were active smokers. Studies included patients with mild to moderate HS, with a mean disease duration of 13.69 years. The HS signs and symptoms assessed were pain, pruritus, malodour and suppuration. Overall, the higher intensity of a sign or symptom correlated with poorer general QoL or specific QoL dimensions including sexual distress, anxiety, depression and sleep. The most frequently employed tool to assess QoL was the Dermatology Life Quality Index (DLQI). DLQI was used in 52.9% of the studies (9/17) with a mean value of 10.70 (2.16 SD). The scores employed to assess signs and symptoms severity were subjective and varied between studies, being the numerical rating scale (NRS) for each of the most used symptoms. The mean NRS value for pain was 3.99 and the mean NRS for pruritus was 4.99. In conclusion, we have summarized, categorized and analyzed the scientific evidence regarding signs and symptoms in HS patients and their impairment in QoL. Their assessment should be thorough and included during routine evaluation of HS patients to motivate therapeutic modifications and increase patients’ health.


Introduction
Hidradenitis suppurativa (HS) is a chronic, recurrent and debilitating inflammatory skin disease of the hair follicle that usually presents after puberty with painful, deep-seated inflamed lesions in the apocrine gland-bearing areas of the body, most commonly the axillae, inguinal and anogenital regions [1,2]. It has an estimated prevalence rate in the

Materials and Methods
A systematic review and meta-analysis were conducted. A literature search was performed using Medline, Scopus and Embase databases from conception to 4 May 2021, following PRISMA Guidelines (Supplementary Material). The following search algorithm was used: (hidradenitis or "acne inversa") and (pain or itch or odour or malodour or suppuration or oozing or drainage) and ("quality of life"). Symptoms included in the literature search were selected by a dermatologist expert in HS (AML) following the most recent evidence in HS clinical presentation [1].
The search was limited to: (i) human data, (ii) articles correlating HS symptoms with quality-of-life impairment in HS patients, (iii) articles written in English. All types of epidemiological studies (clinical trials, cohort studies, case-control studies and crosssectional studies) were included and analyzed. Reviews, guidelines, protocols, case series, case reports and conference abstracts were excluded.
Two researchers (TMV and AML) independently reviewed the titles and abstracts of the articles obtained in the first search to assess relevant studies. The full texts of all articles meeting the inclusion criteria were reviewed, and their bibliographic references were checked for additional sources. The articles considered relevant by both researchers were included in the analysis. Disagreements about inclusion or exclusion of articles were subjected to discussion until a consensus was reached. If not reached, resolution was achieved by discussion with a third researcher (SAS).
The variables assessed were study design, author, country, level of scientific evidence according to the Centre for Evidence-Based Medicine, number of participants, age, sex, BMI (kg/m 2 ), smoking habit, disease duration, disease severity (Hurley stage), HS symptoms and aspects of QoL evaluated, QoL and symptoms assessment tools and scores, correlation between symptoms and QoL.
The mean DLQI and NRS for symptoms was calculated by a random effect metaanalysis weighted by the study sample size. To estimate absolute mean effect of DLQI and NRS for each symptom, the mean, standard deviation and sample size were extracted from the studies. Research with unclear or incomplete reporting was excluded from the meta-analysis. To generate valid estimates, studies were weighed according to their sample size. Forest plots were constructed to assess the distribution of the data and summarize the effect size and their 95% CIs. Quantifying of Heterogeneity was evaluated using Cochrane Q statistic, an intermediary statistic employed to obtain a more useful measure of heterogeneity, the I2. Assuming a high heterogenicity between studies, we used a random effects model to calculate the outcome. Microsoft Excel version 2016, Redmond, Washington, The USA, was used to run this data [22].
The quality of the design was critically appraised using the National Institutes of Health quality assessment tool to evaluate risk of bias (Table S1) [23]. This tool is based on the key concepts for evaluating the internal validity of a study and is divided into a set of 14 set questions. Studies are classified depending on the rate: good quality (>9 criteria met), fair quality (5-9 criteria met) and poor quality (<5 criteria met).
The main characteristics of the studies included are summarized in Table 1. All studies had a cross-sectional design and were classified as scientific level of evidence 4. Samples were recruited from outpatient clinics or through focused electronic, postal or telephone surveys. Study participants were predominantly female. Mean age of the participants was 36.28 years. The BMI of the population was in the range of overweight, about two out five patients were active smokers. Studies included patients with mild to moderate HS, with a mean disease duration of 13.69 years. The body regions more frequently affected by HS were axilla and groins.
The HS signs and symptoms assessed were pain, pruritus, malodour and suppuration. Overall, the higher intensity of a sign or symptom correlated with poorer general QoL or specific QoL dimensions including sexual distress, anxiety, depression and sleep. The most frequently employed tool to assess QoL was the DLQI. DLQI was used in 52.9% of the studies (9/17) with a mean value of 10.70 (2.16 SD) after conducting a random effect meta-analysis weighted by the study sample size (Figure 2
It was found that HS pruritus impaired sleep quality [29,40] and it was linked to poor mental health assessed by MDI [34]. Nevertheless, it was observed that NRS for pruritus did not have an impact on sex life (β = 0.03, p = 0.615) [25], neither in men nor in women [24].
Factors associated with increased risk of pruritus were Hurley III, higher number of regions affected, the female sex, being an active smoker, the intensity of suppuration and pain, having Crohn's disease and not using statins [32,33,40].
. Matusiak et al. observed that the presence of pruritus did not have an impact on QoL [32], while Molina-Leyva et al. observed that the presence of NRS for pruritus > 3 was related with higher rates in DLQI score (β = 0.42 ± 0.11, R2 = 0.20, p < 0.001) [33]. Moreover, higher rates in VAS and NRS for pruritus were positively correlated with DLQI [32,34]. The impact of pruritus in overall QoL was also reflected by Riis et al. showing that higher NRS for pruritus were related to lower values in the EQ-5D (β = −0.017, p < 0.05) [39].
It was found that HS pruritus impaired sleep quality [29,40] and it was linked to poor mental health assessed by MDI [34]. Nevertheless, it was observed that NRS for pruritus did not have an impact on sex life (β = 0.03, p = 0.615) [25], neither in men nor in women [24].
Factors associated with increased risk of pruritus were Hurley III, higher number of regions affected, the female sex, being an active smoker, the intensity of suppuration and pain, having Crohn's disease and not using statins [32,33,40]. Factors associated with increased risk of pruritus were higher BMI, longer disease duration, high number of regions affected and the location on groin, upper thighs, and buttocks, high Hurley stage and intensity of suppuration [33,37].

Suppuration
Three studies evaluated the impact of pain in the QoL of HS patients, including 802 participants with a mean age of 38.83 years [24,25,38], Table 5.

Discussion
The results of this systematic review and meta-analysis presents the clinical situation of patients with HS regarding signs and symptoms and summarizes the current evidence regarding their correlation with QoL impairment, both general and specific. The importance of the research on this topic is notable and increasing, as the majority of the studies are published from 2016 onwards.
As previously described, HS has a great impact on QoL [41], even more than other dermatosis, such as psoriasis or atopic dermatitis [42]. Although most tools used to assess QoL were validated questionnaires, they differed between studies. The tool most frequently used to evaluate QoL was the DLQ, showing moderate to large impacts on patients' lives [26][27][28]30,32,33,35,36]. The scores employed to assess signs and symptom severity were subjective and varied between studies, with the NRS being the most used tool for symptoms [24,32,33]. Furthermore, it is noteworthy that clinicians from North America and Asia are less likely to measure HS symptoms, which may reflect regional differences in clinical assessment or research trends [43].
Pain is the symptom with the strongest correlation with QoL impairment [7,44,45]. Mean pain reported was almost four out of 10, which qualifies as mild-to-moderate pain considering established cut-offs [46]. These values are like chronic posttraumatic headaches and worse than vasculitis, blistering disorders, vulvar lichen sclerosis and leg ulcers [19,47]. HS pain is both nociceptive and neuropathic. Nociceptive pain may be the result of acute inflammation while neuropathic HS pain could be due to chronic inflammation causing peripheral neuroplastic changes and central sensitization. Addressing HS pain is critical to improve HS-related QoL and reduce morbidity from opioid and other substance use. Unfortunately, current HS therapies often provide inadequate pain relief, and studies of HS pain-directed therapies are sparse. Non-steroidal anti-inflammatory drugs, intralesional corticosteroids or neuromodulator medications could be effective treatment for pain [44]. Moreover, incorporation of psychological therapies may represent an important opportunity for reducing chronic HS pain [43]. Pain intensity correlated with impairment in QoL in all the studies included [24][25][26][27][28][29][30][31][32][33][34][35][36]. This is in part explained by the physical limitations caused by the painful lesions. Moreover, pain is associated with poor mental health. Rates of depression and anxiety are higher in HS patients than in healthy individuals [48][49][50]. Pro-inflammatory cytokines, including TNF-α, IL-1β and IL-10, are elevated in the lesional skin of HS patients [49,51]. TNF-α and IL-1β are also increased in major depressive, anxiety and other psychiatric disorders [52]. Therefore, high levels of these cytokines in HS [49,53] could explain the relationship between HS and poor mental health. HS also has an impact on sleep quality, even worse than other systemic conditions, such as lupus erythematosus, chronic obstructive pulmonary disease or Hodgkin's lymphoma [54,55]. Sleep disorders also contribute to decreased QoL [56] and HS pain impact on sleep quality [29]. Sexual health is likewise an important aspect of patients' QoL [57] and pain is also a risk factor for sexual distress and sexual dysfunction [24,25]. Sexual distress reveals the suffering of the subject while sexual dysfunction might mean a poor sexual experience for both members of the relationship [25]. The impact of pain in sexual health may be linked to the nature of the sexual act and psychological factors that may be associated with disease activity [57].
Pruritus is the second symptom with the strongest correlation with poorer QoL [24,25,29,[32][33][34]39,40]. It might be underreported because patients do not spontaneously refer to this symptom unless they are specifically asked [32]. The mean pruritus reported was almost five out 10. Although HS is not considered as a pruritic disease, this symptom is commonly associated with HS, mainly during the outbreak of lesions [20,58]. Pruritus severity has been related to overall impairment of QoL [32][33][34]39], sleep disturbances [29,40] and depression [34] but it has not been linked to poor sexual health [24,25]. In agreement with our results, pruritus has been previously described as a potential risk factor for sleep impairment in other dermatosis [59,60]. The absence of impact on sexual health might be due to pruritus being less bothersome than pain or it could even decrease during sexual intercourse. Skin irritation caused by suppuration might be the cause of pruritus in patients with HS [33]. The increased number of mast cells and inflammatory cell infiltration in HS lesions might also explain the pruritus in HS [7,32]. The reduction of suppuration through antibiotics, anti-inflammatories, or surgical procedures, as well as topical measures to control skin irritation like emollients or corticosteroids, should be considered in patients with the relevant pruritus and suppuration [33].
Scarce studies have evaluated the impact of malodour and suppuration on QoL [24,25,33,[37][38][39]. Compared to pain and pruritus, malodour and suppuration can potentially be perceived by other people and might also contribute to worsen patients QoL and stigmatization. In fact, malodour can be underestimated by the patients as they get used to it, but their partners usually show a more expressive response when they inquire about this problem. This can potentially cause social, work and personal problems, and favours stigma and isolation behaviours [33,61], contributing to decreased QoL. Studies have also shown that malodour and suppuration severity were associated with poorer sexual QoL only in women, not in men [24]. Although previous investigations indicated higher sexual distress in women than in men with HS, the impact of different symptoms between both sexes might be explained by an early onset of HS in women, or even by cultural aspects and differences in emotional and neuroendocrine response to disfigurement [62]. As previously stated, suppuration is linked to advanced chronic lesions and bacterial biofilms [63,64]. Intensity of suppuration, Hurley stage, longer disease duration and high number of regions affected are risk factors for malodour [33]. Structural damage, as in the presence of scars, may make personal hygiene difficult and favour bacterial overgrowth thereby increasing molodour and suppuration scores. Body mass index is associated with malodour probably due to the presence of prominent skin folds and excessive sweating. Control of malodour and suppuration should be a priority in patients with HS to improve their QoL. Weight loss is advisable in all overweight and obese patients. Anti-inflammatories, antibiotics or a combination of both should be given in patients with poor disease control, and antiseptic washes or surgical procedures to remove scarring tissue could be used in patients with good disease control and structural damage [33].
This systematic review is subject to some limitations. All the designs were crosssectional, which limits the inference of causality. Although most of the questionnaires and tools used are validated, there is heterogeneity between the different studies. There are also differences in the severity of the patients depending on the source of the patients (outpatient clinic vs. general surveys). Moreover, some the assessment of these symptoms is subjective, which may also increase the variability.
Further studies should include validating questionnaires to assess QoL and symptoms severity. DLQI might be a good option to evaluate overall QoL. NRS for pain, pruritus, malodour and suppuration should also be included. Outcomes should be reported, both cut-off (nominal) and average (continuous) data of these questionnaires, in future studies.

Conclusions
In conclusion, we have summarized, categorized and analyzed the scientific evidence regarding signs and symptoms in HS patients and their impairment in QoL. Pain might be the symptom most related with impairment in QoL due to its high frequency and subjective component. Malodour is the least studied symptom and could have a major effect on interpersonal relationships. Assessment of these symptoms should be thorough and included during routine evaluation of HS patients. It would be important to define cut-off values of symptom severity to motivate therapeutic modifications. Coordinated and consistent medical and psychological support are of great importance to increase patients' health.

Institutional Review Board Statement:
The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Ethics Committee of Hospital Universitario Virgen de las Nieves, Granada, Spain (protocol code V01 and date of approval 19/05/2019).

Informed Consent Statement:
Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.