ABQOLAutoimmune Bullous Disease Quality of Life Questionnaire

ABSISAutoimmune Bullous Skin Disorder Intensity Score

AIBDAutoimmune blistering diseases

BMZBasement membrane zone

BPDAIBullous Pemphigoid Disease Activity Index

BSABody surface area

COMDQ Chronic Oral Mucosal Disease Questionnaire

DIFDirect immunofluorescence

DLQIDermatology Quality of Life Index

FDAU.S Food and Drug Administration

FDM Fornix depth measurer

ICCIntraclass correlation coefficient

IGAInvestigator’s Global Assessment

IOCIntraoral camera

MMPMucous Membrane Pemphigoid

MMPDAIMucous Membrane Pemphigoid Disease Activity Index

OcMMP Ocular MMP

ODSSOral Disease Severity Score

OHIPOral Health Impact Profile

PDAIPemphigus Disease Activity Index

RCTRandomised controlled trial

TABQOLTreatment of Autoimmune Bullous Disease Quality of Life Questionnaire

UATUpper aerodigestive tract

One of the earliest global scoring systems used for MMP is the unnamed multisite MMP score proposed by Setterfield [5]. This score is still occasionally used due to its emphasis on oral involvement though it has not been validated [22]. The Setterfield score is applicable to all MMP types, including severe cases of MMP, and heavily weights high risk areas and mucous membranes (only 14% of the score is for skin sites) [2]. The score has also been found to correlate with IgA, IgG and BP180 reactivity; patients with both IgA and IgG on the BMZDIF and/or low BP180 reactivity correlated with high clinical scores [5, 22]. Despite this the score by Setterfield [5] lacks a distinct separation of activity and damage scores, which means a high score could indicate either a patient with active disease or a patient with controlled disease but severe complications. The score also includes subjective measurements such as mild, moderate, and severe in the larynx column.

The MMPDAI is a specific scoring system developed by an international panel of dermatologists who specialise in AIBD and is based on the existing and validated PDAI and BPDAI scores [4]. Whilst not validated in its entirety, the score is recommended for use in MMP by the German S2k Guidelines and European S3 Guidelines [2, 7].

The weighting of the scalp (4%), mucous membranes (48%) and skin (48%) have been altered in comparison to the Pemphigus Disease Activity Index (PDAI) and Bullous Pemphigoid Disease Activity Index (BPDAI) (10% and 45% respectively) to better represent the distribution of MMP [4]. The score also features a damage column to quantify post inflammatory changes and scarring, which most commonly cause complications in MMP [4]. Other merits include the ability to score the eyes without an ophthalmologist, and good intra-rater reliability [4, 11, 12]. The authors of the MMPDAI proposed an addendum in 2015 to account for the lack of a descriptor for lesions that are stable but erythematous to indicate activity [4]. The oral component of the score has been validated [11].

The MMPDAI has shown promising correlation with the pathological picture of MMP. Corti [23] found that MMPDAI score was associated with circulating IgA and linear IgE deposits on the BMZ and Endo [14] found that a high oral disease activity MMPDAI score correlated with upper aerodigestive tract (UAT) involvement. A case report of a patient with an oral disease MMPDAI score of 44, who developed acute respiratory failure secondary to laryngopharyngeal stenosis, also confirms this pattern [21]. Whilst individual case reports are a weak form of evidence, they are useful in reinforcing the results of Endo [14]. These are promising results as they suggest that an increased MMPDAI score correlates with high risk MMP, and thus a patient’s potential for life threatening laryngeal scarring and stenosis. Overall, the papers by Endo [14] and Corti [23] have similar limitations: small sample sizes and the retrospective application of the MMPDAI to patient notes. It is not specified if the raters were aware of the severity of patients’ MMP or if there were any serious complications which may influence their rating. Specifically, Endo [14]’s work fails to examine the correlations of MMPDAI scores with UAT in asymptomatic lesions and Corti [23] is limited by the low sensitivity of indirect immunofluorescence. It is also not specified if the reported scores in any of the above studies are combined activity and damage scores, or only activity scores.

Whilst the MMPDAI is one of the only generalised disease specific tools that can be used by members of a multidisciplinary clinical trial team, it is only proposed for mild MMP [2, 4]. Unlike in the Setterfield score, it is recommended that specific scales be used to quantify the degree of ocular, oral, laryngeal, esophageal or pharyngeal mucosa involvement in higher risk cases of MMP as this cannot be measured in detail by the MMPDAI [4]. Corti [23] had to estimate a patient with only esophageal involvement’s MMPDAI as 10 and Ormond [11] found the oral sections of MMPDAI had poor interrater reliability for damage scores (prior to the erythema addendum). The MMPDAI could also benefit from review by oral medicine physicians, otolaryngologists and ophthalmologists with specialist knowledge of the effects of MMP outside of the skin, which may enable the score to be applicable to more specific cases of MMP.

The Autoimmune Bullous Skin Disorder Intensity Score created by Pfütze [24] is recommended by the European S3 guidelines for MMP [2, 24]. It is a score reflecting clinical disease activity with a specific section for the assessment of oral mucous membranes [7, 11, 24]. There is also a subjective component in oral assessment which assesses a patients discomfort when eating and drinking [11, 24]. The types of foods in the subjective component can be swapped with foods that are preferred regionally but are texturally equivalent to account for differences in patients’ diets [24]. The ABSIS has been validated for oral MMP and found to have good interobserver reliability in oral MMP. This was better than the MMPDAI activity score which only showed satisfactory correlation, but not as good as the ODSS [11]. This score also uses the rule of nines to calculate body surface area (BSA) affected. Whilst the rule of nines is a quick tool to estimate BSA it often overestimates the area affected [25]. Rosenbach [26] found the ABSIS poorly represents small changes in disease activity at the low end of the disease spectrum in pemphigus. Although this study examined patients with pemphigus, a similar AIBD, it did with not include patients with MMP. Therefore examining the validity and reliability of the ABSIS in MMP would be useful. A flaw of using the ABSIS for MMP is that it lacks a score for ocular involvement, which the authors propose can be combatted by using the Tauber score in conjunction [24, 27]. Despite this the ABSIS is not disease specific and has not been validated in multisite MMP [2].

The oral disease severity score (ODSS) is a score used for chronic oral diseases that has been validated in MMP [11]. The ODSS is recommended for use in mono-site and predominantly oral MMP by both the European S3 Guidelines and the German S2k guidelines for MMP [2, 7]. In the ODSS, which has previously been used in lichen planus and pemphigus vulgaris, 17 areas of the oral cavity are given a site score, area score and pain score which are then added to give a maximum score of 106 [11, 12]. This is more detailed compared to the division of the oral cavity into 7 areas in the MMPDAI and 11 in the ABSIS [4, 11].

Potential weaknesses of the ODSS include the combination of the subjective pain score and objective site and activity scores, which the authors of the ODSS feel provides a more comprehensive representation of disease experience [11]. This subjective component also does not represent a quality-of-life measure [11]. The score lacks a damage component, which despite the decreased propensity of the oral area for scarring is still concerning as most complications of MMP are linked to post inflammatory changes and scarring [1].

The ODSS was validated in a single point-of-time study of 15 patients with mild to moderate MMP, during which ten physicians scored each patient, and two physicians rescored their patients to test intra-rater reliability [11]. It was found the ODSS had good convergent validity with the ABSIS and MMPDAI and an intra-rater reliability intraclass coefficient (ICC) of 0.97 and 0.93. The study made many efforts to reduce bias, such as having a two hour wait time before patients were rescored and using specialists from multiple fields of medicine such as dermatologists and oral health specialists. Despite this, limitations include that none of the clinicians regularly used the ABSIS or MMPDAI, and only half were experienced using the ODSS.

Methods of scoring laryngeal MMP have been proposed by both Nash [34] and Higgins [15] [2, 15, 34]. The Higgins scoring system is scored based on lesion type, site and airway patency, combining activity and damage stages [15]. In comparison the Nash scoring system is based on symptomology, due to the authors concerns that it would be difficulty to routinely assess the larynx [34]. Whilst both scores exist neither have been validated and it is instead recommended that an otolaryngological version of the MMPDAI be used for clinical studies [2]. In pemphigus vulgaris a score combining number of lesions and lesion size has been used, which was shown to correlate with the PDAI, ABSIS and endoscopic examination [35].

The Foster scoring system, followed by the Mondino and adapted Tauber tool are the most common scoring tools used in ocular MMP [8, 11]. Foster’s score divides OcMMP into four stages based on general signs of cicatrisation whilst Mondino’s scoring system only uses inferior fornix shortening [16, 36]. Whilst this score is more objective than the Foster score, it lacks sensitivity to early disease stages [37]. The Foster score is insensitive to inferior fornix loss or increases in symblepharons and the Mondino score fails to represent new symblepharons in the absence of fornix loss [37].

The Tauber scoring system was created to combat insensitivities in the Foster and Mondino scores [27]. This score follows the Foster stages but incorporates a sub score for inferior fornix shortening and horizontal symblepharon involvement [37]. The Tauber score was validated in a retrospective cohort study of 125 eyes against the Foster and Mondino scores. All scores showed excellent interrater reliability, with the Tauber score best representing if MMP had advanced [37]. A systematic review and meta-analysis by Bocanegra-Oyola [8] recommend the Tauber system as it combines the variables in the Foster and Mondino scores, making it significantly more accurate.

The Foster, Mondino and Tauber tools can all be used retrospectively on clinical records due to their simplicity [29, 38]. This is particularly advantageous in rare disease such as MMP where most studies are retrospective. Stage IV of the Foster score has also been found to have a strong correlation with poor visual acuity (>6/60), whilst the Mondino score has been shown to have a poor correlation [16].

In a literature review of all available scoring systems for cicatricial conjunctivitis Ong [29] noted that the Foster, Mondino and Tauber tools overlook upper fornix scarring, which is also associated with complications threatening sight, and are largely qualitative measures based on a physicians clinical judgement. Other weaknesses include that the Foster, Tauber and Mondino scores only assess scarring, making them insensitive in detecting disease activity, and that approximation of inferior fornix shortening requires visualisation of the posterior lid surface [28, 38]. These scores also require lid traction, which can cause distortion of the lid which alters final scoring [28]. Despite this useful correlation, visual acuity is a poor measure of OcMMP progression as it cannot be assumed to be due to OcMMP as can instead be caused by a patients pre-existing comorbidities [38].

International guidelines recommend the validated cicatrising conjunctivitis assessment tool (CCAT) be used for ocular MMP [2, 7]. The CCAT validated by Ong [38] is composed of an inflammation, scarring and morbidity score. Unlike other scores discussed above the score assesses both the upper and lower fornix [38]. Similarly to the MMPDAI, the score divides the bulbar conjunctiva into four quadrants to better approximate conjunctival inflammation. All components of the assessment tool were individually validated, and only kept in the score if adequate agreement was shown [29]. The scarring and morbidity components, which are designed to represent the irreversible effects of the disease, had excellent intra-rater and interrater correlation [38]. The percentage of lower fornix shortening is also calculated using a FDM and age standardised fornix depth tables to provide a much more objective result than simply visualising fornix loss [27, 29]. Despite these strengths the tool still requires specialist equipment such as slight lamps and FDMs and only provides fornix depth tables for Caucasian populations. Since the CCAT validated in a mainly Caucasian population, further validation studies are required for diverse populations

In the CCAT, reference images are used to aid in scoring of conjunctival hyperemia, which is in line with current expert recommendations [6, 7, 38]. The creation of a visual scale to document hyperemia was first proposed Elder and Bernauer [16] in criticism of the Foster and Mondino scores. The detection of active inflammation is crucial in OcMMP as this allows the disease to be treated before vision threatening conjunctival scarring develops [29]. Despite this, Saw [39] showed progression of cicatrisation wasn’t always associated with less severe inflammation. It is thought these results occur because persistent mild inflammation in a patient with treatment resistant disease may cause cicatrisation to progress [39].

Multiple other scoring systems have been proposed for OcMMP such as the score by Reeves [28] and Rowsey [40] but these have been excluded from this review due to infrequent use [28, 29, 40].

The authors have declared that no competing interests exist.

The authors declared that no clinical trials were used in the present study.

The authors declared that no experiments on humans or human tissues were performed for the present study.

Informed consent from the humans, donors or donors’ representatives: Bellberry Limited.

The authors declared that no experiments on animals were performed for the present study.

The authors declared that no commercially available immortalised human and animal cell lines were used in the present study.

No use of AI was reported.

No funding was reported.

Conceptualization: DM. Writing – original draft: AH. Writing – review and editing: DM, MS, AH.

Annabelle Heyworth https://orcid.org/0009-0001-9332-7365

All of the data that support the findings of this study are available in the main text.