Anemia in patients with hidradenitis suppurativa

Hidradenitis suppurativa (HS) is associated with a number of physical and psychological comorbidities. Studies have suggested an association between HS and anemia; however, this association is not widely understood and may result in delayed diagnosis and treatment and possible increase in morbidity and mortality. To systematically review and perform a meta-analysis regarding the association between HS and anemia, and to characterize the subtypes of anemia associated with HS. A search of the EMBASE, Medline, Web of Science Core Collection, and Cochrane Central Register of Controlled Trials databases from the time of database inception to September 25, 2022, yielded 313 unique articles. All observational studies and randomized controlled trials published in English that examined the odds ratio (OR) of anemia in patients with HS were screened by 2 independent reviewers. Case reports were excluded. Among 313 unique articles, 7 were deemed eligible. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines facilitated data extraction. The Newcastle–Ottawa Scale (NOS) was used to analyze risk of bias of included studies. In addition to OR and 95% confidence intervals, relevant data on patient demographics and anemia subtypes were also extracted. The primary outcome was the OR of anemia in HS patients. This study also attempted to characterize anemia subtypes associated with HS. In total, 2 case–control and 5 cross-sectional studies featured a total of 11,693 patients. Among the studies, 4 of 7 demonstrated a statistically significant positive association between anemia and HS (ORs, 2.20 [1.42–3.41], 2.33 [1.99–2.73], 1.87 [1.02–3.44], and 1.50 [1.43–1.57]), with macrocytic and microcytic subtypes being most common. After adjusting for publication bias, meta-analysis with random effects revealed HS to be associated with increased odds of anemia compared to non-HS groups (OR 1.59 [1.19, 2.11]). There is a statistically significant positive association between HS and anemia, particularly for the aforementioned subtypes. Patients with HS should be screened for anemia. In case of lower hemoglobin concentration, the anemia of HS patients should be subdivided according to mean corpuscular volume of the red blood cells and further investigated depending on subtype.

The association between HS and anemia as a potential comorbidity has been explored but is unclear to date. Anemia is the most common hematologic condition in the U.S., affecting approximately 6% of the population [24] and is defined as having a hemoglobin (Hb) concentration below 13 g/dl in men, below 12 g/dl in nonpregnant women, and below 11 g/dl in pregnant women [25,26]. Anemia is characterized by various subtypes including iron deficiency, folate deficiency, pernicious, sickle-cell, hemolytic, and chronic disease [27]. Among the subtypes, anemia of chronic disease has been the most common subtype as being associated with chronic inflammatory diseases like HS [28,29].
Some studies, especially those attempting to analyze comorbidities of HS, have demonstrated a positive association between anemia and HS [30][31][32][33][34]. However, the association between the two conditions is not yet conclusive, as other studies have indicated lack of an association [35]. Furthermore, studies to date are of varying methodological designs, with study samples bearing different demographics [30][31][32][33][34][35]. Therefore, in order to better understand the association between HS and anemia, we systematically reviewed the existing literature and performed a metanalysis, paying special attention given to the methodological quality of relevant studies. We also sought to characterize the subtypes of anemia associated with HS.

Methods
Our systematic review was conducted in accordance with the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) [36]. Two independent reviewers (R.R.R. and A.P.) determined eligibility criteria, screened titles and abstracts, reviewed full articles, extracted data, and evaluated eligible studies for methodological quality. Two reviewers (R.R.R. and A.P.) made final decisions on eligibility after fullarticle review using Covidence. When the full version of an article was not available online, corresponding authors were successfully contacted to provide access.

Search strategy
A comprehensive search query that combined relevant key words and subject headings, hidradenitis suppurativa, inverse acne, Velpeau's disease, anemia, and hemoglobin was developed by an experienced medical librarian (S.W.). We searched the following databases: Ovid EMBASE, Ovid Medline, Web of Science Core Collection, and Cochrane Central Register of Controlled Trials. The specific search query is available in the eAppendix section of the Supplement. Eligible studies were manually screened to determine whether their list of references presented any additional articles to consider for inclusion.

Eligibility criteria
All articles from database inception to the date of search (September 25, 2022) were considered. Only studies published in English were considered. Randomized controlled trials and observational studies (case-control, cross-sectional, and cohort studies) were considered; however, case reports were excluded. To be considered for inclusion, studies were required to include a group of HS patients that was statistically compared to a non-HS group of patients. Furthermore, these studies had to present either unadjusted or adjusted odds ratios (ORs) for anemia in patients with HS. Diagnosis of HS was determined as per the following: (1) physician-verified diagnosis, (2) International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) code 705.83, or (3) national surveys with predetermined sensitivity and specificity for identifying HS [35]. Diagnosis of anemia was determined as per the following: (1) clinical diagnosis based on World Health Organization guidelines as hemoglobin (Hb) concentration below 13 g/dl in men, below 12 g/dl in non-pregnant women, and below 11 g/dl in pregnant women, (2) ICD-9-CM code 285.9, or (3) identification by medical claims recorded during the study period. The inclusion flow chart is shown in Fig. 1.

Data extraction
The primary data extracted were prevalence, unadjusted and adjusted ORs, 95% confidence intervals, and p values. Variables controlled for adjusted analyses include age, sex, race/ethnicity, smoking status, insurance status, alcohol use, BMI, ALT, and HbA1C. If not explicitly presented, unadjusted ORs were manually derived using the given prevalence data for anemia in HS and comparison groups. In addition to ORs, we also obtained the following data: (1) first author, (2) setting-time period under study, location, outpatient vs. inpatient, (3) study design, (4) sample size, (5) population demographics-sex and age (7) methods of diagnosing HS and anemia, and (8) severity assessment for HS.

Risk of bias assessment
To evaluate bias, we used the Newcastle-Ottawa Scale (NOS) [37] which was developed to assess the quality of nonrandomized studies. The original NOS was designed for quality assessment of case-control and cohort studies; later a version was adapted for cross-sectional studies [14,38,39]. Case-control studies were evaluated on a scale of 0-9 for criteria on sample selection, comparability and exposure, while the cross-sectional scale ranged from 0 to 10 for criteria on selection, comparability and outcome. Higher scores indicated better quality studies. The scale has face validity, criterion, and construct validity. Though interrater reliability and threshold scores are in development, we adopted the following thresholds for the NOS scores [14,38,39]: "good quality" (3-4 points in the "selection domain" AND 1-2 points in the "comparability" domain AND 2-3 point in the "outcome" domain), "fair quality" (2 points in the "selection domain" AND 1-2 points in the "comparability" domain AND 2-3 points in the "outcome" domain), and "poor quality" (0-1 points in the "selection domain" or 0 points in the "comparability" domain or 0-1 point in the "outcome" domain).
Studies were evaluated independently by 3 reviewers (R.R.R., A.P., and M.R.). Disagreement was resolved by discussion. A figure summarizing our methodological assessment may be found in the Supplement. For each category of the NOS, a rating of "low risk of bias" was denoted by a star, and "unclear risk of bias" by a question mark. For all studies, we awarded 1 star for comparability of cases and controls if controls were matched based on sex and age at the time of assignment or analysis. Two stars were awarded if controls were matched based on sex, age, and BMI and/ or smoking. Publication bias was assessed by the visual inspection of the funnel plot.

Statistical analysis
RevMan 5.4.1 was used to conduct the meta-analysis using the methodological review type. Event and totals for HS and anemia were obtained from the studies. RevMan calculated the ORs and 95% CI using a fixed effect model. These values were visually compared to the published values, when available to validate the calculations. For the analysis on the adjusted ORs, we used the generic inverse variance method to obtain the pooled effects. When any parameters were missing, we used the RevMan calculator to estimate necessary variance outcomes. Cochran's Q test and I 2 statistics were used to assess the heterogeneity among studies, and I 2 > 50% indicates significant heterogeneity.

Risk of bias
With regards to case-control studies [30,32], the mean (standard deviation [STDEV]) NOS was 5.5 (3.5). For Kimball et al. unclear risk of bias was rated for "adequacy of case definition" due to the use of ICD-9-CM codes as a form of record linkage without independent validation [37]. "Non-response rate" was also rated as unclear risk of bias although it was mentioned that cases of low severity may have gone untreated, it was not possible to estimate how many of such cases existed within the large data set. Ponikowska et al. received the rating of unclear risk of bias for several categories. Although it was mentioned that a control group of healthy patients was used for comparison, absence of details regarding screening/eligibility criteria made it impossible to assess "comparability of cases and controls." Furthermore, "ascertainment of exposure was unclear;" while the Hidradenitis Suppurativa Severity Index and Sartorius scale were mentioned, no specific information was present regarding the confirmation of diagnosis through patient interviews or medical records. With regards to cross-sectional studies [31,[33][34][35], the mean (STDEV) NOS was 7.38 (2.58 (2) for the population HS group. The "sample size" for cross-sectional studies was rated with "unclear risk of bias" when implicitly large but not quantified [31] or when quantified but not justified as to why a given number of cases were selected [34]. Studies with samples of relatively smaller size were also rated similarly [35]; however, unlike other studies with unclear bias, Miller et al. (1), was awarded a star for "representativeness of sample," because the participation rate for all invited cases was stated (34%), and distribution of age and sex did not differ between participants and non-participants. This study was also the only cross-sectional study to report this data on "non-respondents." Finally, studies that implied the use of medical records or ICD codes for record linkage to databases but that did not explicitly state so received the rating of unclear risk of bias for "ascertainment of exposure" [31,34].

Hidradenitis suppurativa and anemia
Six out of 7 included studies presented either unadjusted ORs or prevalence data that allowed for calculation of unadjusted ORs [30,[32][33][34][35] (Fig. 2). Based on unadjusted ORs, 3 out of 6 potential studies demonstrated statistically significant positive associations between HS and anemia [30,33,34]. After pooled estimation, it was found that HS patients were associated with a significantly increased odds  [31,34]. The other 2 studies (Miller et al. 1 and 2) presented adjusted ORs demonstrating no association [35]. After pooled estimation, we found HS was associated with a significantly increased odds of anemia compared to the comparison groups (OR 1.59 [1.19, 2.11]).

Reporting bias
The funnel plot (Fig. 3) displays the evaluation of reporting bias. Three of the seven studies fall at the apex of the triangle due to large sample sizes (weights) and are the only ones that are statistically significant (Ramos et al., Kimball et al. and Solimon et al.). One study fell on the negative side (though it was not statistically significant, Miller et al. (1)). There does appear to be a publication bias with the studies included as the plot is asymmetrical and lacks small and negative studies. The small number of studies serves as a limitation and heterogeneity of the studies was also considerable, I 2 = 91%.

Discussion
To our knowledge, this is the first systematic review and meta-analysis conducted on the association of HS and anemia, while paying particular attention to the various subtypes of anemia in each study.
Four out of 7 studies demonstrated a positive association between anemia and HS. This is consistent with studies in the literature, which report prevalence of anemia up to 40% in demographically diverse HS patient cohorts as compared to approximately 6% in the general population [25,29,40]. Studies demonstrating a positive association indicated prevalence of anemia in HS patient groups to be significantly higher than that in the general population. Considering both HS and anemia being more common in females of childbearing age [5][6][7], Soliman et al., and Narla S., Silverberg J.I., adjusted for sex, age, and other potential confounders, and still found a statistically significant association [25,41]. Both studies conducted by Miller et al. revealed no association and adjusted analysis determined no statistically significant results. This finding may be reflective of the difference in population demographics: whereas HS is most prevalent in black females [5][6][7], the study samples For each study, the samples were characterized as either "inpatient," "outpatient," or "population." All studies assessing inpatient samples revealed a positive association [30,31,33,34]. Among those assessing outpatient samples, 1 out of 2 revealed a positive association [32,35]. Among population studies, 0 out of 1 revealed a positive association [35]. Among studies that conducted HS severity assessments (Table 1), mixed results were found regarding the role of HS severity in anemia. Both studies by Miller et al. showed a lack of association. Similarly, while reporting on microcytic anemia, Ponikowska et al. found that the prevalence of iron deficiency in HS patients was not related to HS severity. In contrast, Kimball et al. noted higher prevalence of anemia in patients with severe HS (11.1%) vs. mild HS (8.0%). Likewise, Soliman et al. reported that patients with severe HS had lower Hb levels and were 6 times more likely to be anemic than patients with mild HS. Using multivariate analysis, a retrospective cohort study comparing comorbidities in patients with mild vs. severe HS found a positive association between severe HS and anemia (OR 23.31, 95% CI 1.73-313.5) [42]. Future studies are needed to clarify the association between HS severity and anemia.
Data from each study regarding subtypes of anemia varied significantly. Kimball et al. found a positive association with "deficiency anemias" as a whole. Associations with microcytic anemia were noted by Soliman et al. and Ponikowska et al. In the latter study, HS patients were found to have lower levels of iron, transferrin saturation (Tsat), ferritin and hepcidin (p < 0.001). Despite indicating no association between HS and anemia, both studies by Miller et al. revealed that among HS patients with anemia, 60% had normocytic anemia and 40% had microcytic anemia. Furthermore, macrocytic anemia was diagnosed in 1 out of 2 patients with anemia in the hospital case group but in none of the 106 population cases with anemia [35]. Studies that report on population samples with greater demographic diversity and in areas where anemia is more prevalent are needed to confirm these findings. Narla S., Silverberg J.I., reported a positive association with pernicious anemia [43], an autoimmune condition characterized by autoantibodies that attack gastric parietal cells, which produce the intrinsic factor necessary for the proper utilization of vitamin B12. In comparison, HS is an autoinflammatory disease, with well-established associations to other autoimmune diseases like inflammatory bowel disease (IBD) [33,44,45]. Therefore, future studies that take into consideration demographic data are needed to validate the association between the autoimmune and autoinflammatory conditions of pernicious anemia and HS, respectively.
The correlation of microcytic and normocytic subtypes is consistent with attributing anemia in HS patients to anemia of chronic inflammation [28,29]. Iron is involved in mechanisms integral to the pathogenesis of chronic inflammatory diseases [46,47]. In such cases, the peptide hormone hepcidin is increased, preventing the release of iron from ferritin stores, resulting in microcytic anemia. Conversely, when the iron stores themselves are depleted, such as in iron deficiency anemia, low hepcidin levels may occur [32,48]. With chronic inflammatory diseases such as HS, anemia can result over time due to elevated hepcidin levels. However, iron deficiency anemia is also prevalent in the women of childbearing age, who are disproportionately affected by HS. Hence, future studies are advised to record hepcidin levels in the analysis of microcytic anemia, as in the case of Ponikowska et al.
Beyond the subtypes of anemia associated with chronic inflammation, others, including sickle cell and pernicious subtypes, merit further study in order to better understand additional mechanisms by which HS may affect red blood cells. Studies in the past have found statistically significant associations between HS and sickle cell anemia [29]. However, our list of included studies did not address this subtype. Future studies including demographic data are needed to better characterize this association, given that HS and sickle cell anemia are both more common in African Americans [49].

Limitations
Limitations to this study also deserve consideration. All 7 studies included for systematic review were conducted in the West (Europe and the U.S.) [30][31][32][33][34][35]. Because all studies were observational, causality was not assessed. Patients with HS are potentially higher healthcare users in comparison with healthy controls, which could lead to the increased likelihood of having a complete blood count (CBC) ordered. Therefore, the studies included may present patient samples that reflect an element of detection bias. Furthermore, there were no cohort studies that met criteria for inclusion. Varying methodological quality and control groups existed among included studies. Not all studies adjusted for potential cofounders in OR calculations. The importance of doing so becomes evident when considering the association between HS and comorbidities like IBD, which has revealed that HS patients are more likely to have IBD than the general population [33,44]. Incidentally, patients with IBD are also more likely to be anemic, which may lead to confounding [50]. Future studies that control for potential confounders like this one are needed to better understand the association between HS and the various subtypes of anemia.

Conclusions
The evidence overall supports a positive association between HS and anemia [30][31][32][33][34][35]. This finding should be validated by further research efforts that adjust for potential confounding variables and seek to identify causal relationships.
Increasing awareness regarding the association between anemia and HS is critical because many patients with anemia are asymptomatic and could otherwise be left undiagnosed, resulting in significant morbidity and mortality [27]. Our data suggests that patients with HS should be screened for anemia; in case of lower hemoglobin concentration, the anemia of HS patients should be subdivided according to mean corpuscular volume of the red blood cells and further investigated depending on subtype. Our results, furthermore, raises an important question: do any existing treatments for HS serve to improve anemia symptoms? Future studies are needed to investigate anemia outcomes with different treatments in demographically diverse cohorts.
Author contributions AP and RR cultivated the ideas underlying the research question.AP, MR, SW, and RR performed the database search, data extraction, and methodological quality analysis.MR, RR, and HP performed the meta-analysis.MR, RR, HP, and NR prepared table and figures.AP, RR, HP, and JH wrote the manuscript.RR, NR and JH prepared the submission materials.All authors reviewed the manuscript.
Funding Authors did not received any funding for this work.
Data availability My manuscript has data as included in electronic supplementary material.