A “modified Obel” method for the severity scoring of (endocrinopathic) equine laminitis

Introduction

Laminitis is a debilitating condition affecting all equids and is associated with several systemic diseases that may result in structural changes within the horse’s foot (Pollitt, 2016). It results in lameness, morbidity and mortality (Patterson-Kane, Karikoski & McGowan, 2018). Laminitis creates attrition within the horse industry with a recent study finding that 33% of all cases were humanely destroyed within 12 months of diagnosis (Luthersson et al., 2017).

Recently, endocrinopathic laminitis, associated with equine metabolic syndrome and pituitary pars intermedia dysfunction, has been acknowledged as the most prevalent form of the disease occurring in horses and ponies (Donaldson, Jorgensen & Beech, 2004; Karikoski et al., 2011; Patterson-Kane, Karikoski & McGowan, 2018). All forms of laminitis share the common clinical sign of foot pain, although it is recognized that the endocrinopathic form may have a milder clinical presentation, and be more insidious in onset than other forms of laminitis (McGowan, 2008; Pollitt, 2004). This is supported by histopathological evidence of laminitis occurring prior to the clinically apparent presentation of disease (Karikoski et al., 2015; Kawasako et al., 2009; Morgan et al., 2003).

The most widely accepted tool for diagnosing and categorizing the severity of laminitis is the Obel method, which was developed 70 years ago using cases of sepsis-related laminitis (Obel, 1948). The Obel method is favored in clinical research, having been used across a range of laminitis induction studies including the oligofructose method (Dern et al., 2017; van Eps & Pollitt, 2006), the prolonged euglycemic hyperinsulinemic clamp (Asplin et al., 2007; de Laat et al., 2010), and most recently, in a dietary model using non-structural carbohydrates (Meier et al., 2018a). Thus, the Obel method is applicable across several forms of laminitis with moderate repeatability and substantial reproducibility as demonstrated in one study of 25 ponies with acute pasture-associated laminitis (Menzies-Gow et al., 2010). Additionally, the Obel method has been compared to the clinical grading system (CGS) for evaluating equine lameness, and the visual analogue scale (VAS) which defines lameness severity along a 10 cm line according to a pain assessment from sound (clinically normal) to the worst possible pain (Viñuela-Fernández et al., 2011). Although these methods demonstrated high reliability, neither the CGS nor the VAS described the laminitis gait or allocated a laminitis score or grade. It’s also important to note, any laminitis grading system is assessing clinical foot pain, as opposed to the severity of the lamellar lesion. Laminitis pain has also been assessed using the Horse Grimace Scale which interprets facial-expressions, removing the need for locomotion; although promising results were seen in this study, the authors recommended further validation before clinical application (Dalla Costa et al., 2016).

Nevertheless, laminitis researchers are aware that diagnosing laminitis can be difficult due to the non-specific nature of clinical signs and lack of robust case definitions (Wylie et al., 2016). In particular, the difficulty in diagnosing mild cases of endocrinopathic laminitis was brought to light in our recent induction study (Meier et al., 2018a). A total of 37 ponies were fed a high non-structural carbohydrate diet over 18 days to induce laminitis. The insulin concentrations in response to this diet, measured over 4 h, were later used to determine an insulin risk threshold for the development of laminitis on this challenge diet. However, 10/11 ponies with insulin concentrations that exceeded the defined threshold for laminitis risk in the study displayed some clinical signs of laminitis such as a short, stilted gait, weight shifting and/or an increased digital pulse, but were not severe enough to meet the criteria for an Obel grade 1 diagnosis, as judged by the two laminitis experts (Meier et al., 2018a). Furthermore, this study found that cases of endocrinopathic laminitis did not always fit clearly within the individual Obel grades, with symptoms that often spread over two grades.

These findings highlighted that a modification of the Obel method would be useful to allow for scoring of individual clinical signs, with a graduated scale that could identify small changes over time. This could assist laminitis researchers to test treatment and preventive strategies, an important requirement as research progresses toward elucidating laminitis pathophysiology and developing efficacious therapies (Frank, 2018; Lane et al., 2017; Legere et al., 2018; Meier et al., 2018b).

Hence the aim of this study was to modify the Obel method of laminitis diagnosis into an examination that can characterize the severity of specific clinical signs, enabling the determination of a laminitis severity score. The modification needed to be suitable for use in monitoring the rate of recovery from laminitis in clinical trials assessing laminitis treatments.

Materials and Methods

Results

Agreement between Obel and “modified Obel” methods

The refined “modified Obel” method and the Obel system were employed by 28 veterinarians to grade laminitis severity in 15 video recordings, with the results presented in Table 4. There was a clear agreement between the two methods when the “modified Obel” scores were converted to Obel grades. As shown in Fig. 1, the methods gave identical results in 62% (259/420) of scorings. In 35% (148/420) of scorings there was a difference of only one grade between the two methods, and a disparity of two grades found in only 3% of scorings (13/420). This result demonstrates a strong relationship between the two systems. Additionally, the mean difference (95% CI) between the 420 scorings showed that overall, the Obel method scored only −0.12 (−0.19 to −0.06) grades lower than the “modified Obel” method.

Table 4:

Weighted kappa statistics (kw) for the QUT and Obel methods determining the intra-observer agreement (repeatability) for 14 veterinarians (raters) who graded a series of 15 laminitis videos twice.

	“modified Obel” method	P-value	Obel method	P-value
Rater 1	0.703	0.0059	1	0.0001
Rater 2	0.989	0.0001	1	0.0001
Rater 3	0.879	0.0005	0.812	0.0014
Rater 4	0.743	0.0011	0.86	0.0006
Rater 5	0.994	0.0001	1	0.0001
Rater 6	0.735	0.0042	0.905	0.0003
Rater 7	0.707	0.0053	0.931	0.0003
Rater 8	0.885	0.0005	0.907	0.0004
Rater 9	0.919	0.0004	0.814	0.0016
Rater 10	0.675	0.0083	0.911	0.0004
Rater 11	0.693	0.0068	0.842	0.0009
Rater 12	0.932	0.0003	0.905	0.0004
Rater 13	0.612	0.0171	0.925	0.0003
Rater 14	0.712	0.0054	0.894	0.0003
Average kw	0.80		0.91

DOI: 10.7717/peerj.7084/table-4

Discussion

Overall, this study showed that both the Obel method and the new “modified Obel” method had excellent agreement, reproducibility and repeatability, when tested by 28 experienced equine veterinarians using 15 video recordings of endocrinopathic laminitis of varying severity (and normal animals). The Obel method was modified by using and analyzing a prototype method in a recent laminitis induction study, studying recent relevant literature and consulting expert opinion. The strength of the “modified Obel” method lies in its ability to score the severity of common clinical indicators of laminitis, which when combined, equate to an overall severity score (Table 2).

Importantly, in using the “modified” Obel method we do not prescribe a score or a level at which “clinical laminitis” is diagnosed. Although excellent agreement was found when the “modified Obel” method was converted to an Obel grade, the authors recommend using it as a stand-alone severity-scoring method, for example, in clinical trials monitoring laminitis recovery. A diagnosis of endocrinopathic laminitis needs to be made by a veterinary professional in conjunction with clinical acumen and supporting evidence such as history, as well as physical examination findings.

Critical to the development of this “modified Obel” method was the careful selection of the five-criteria included in the three-stage examination that would best capture all cases of endocrinopathic laminitis (Table 2). In examining the first criterion of weight shifting, our previous study found that the classic “alternating forelimb” weight shifting captured many, but not all of the laminitis cases (Meier et al., 2018a). Instead, other forms of weight shifting were seen as well, including lying down, and what is known as the “laminitis stance” where the forelimbs are placed in front of the body so as to bear more weight on the hindlimbs (Swanson, 1999), or to ease tension on the deep digital flexor tendons. Although this laminitis stance was seen in less than half of animals observed by Wylie et al. (2013), a further study found that animals that were not “shifting weight” between the forelimbs but had the “front feet placed in front of the body” were identified 94.2% of the time as laminitis cases (Wylie et al., 2016). Hence, expanding the description of weight shifting was necessary.

Lifting a forelimb by an operator was another contentious criterion, as it was argued on one side by some laminitis experts that the lift had to be difficult in each forelimb, as laminitis (with the exception of the supporting-limb form) is a bilateral disease. There is strong supporting evidence for this statement, as the laminitic gait at the walk is bilaterally stilted, and 92% of animals with this clinical sign did have laminitis (Wylie et al., 2016). Furthermore, in a previous study, laminitis was reported to occur in one forefoot in only 3.6% of cases (20/553), whereas it occurred in both forefeet 53.5% (296/553) of the time and in all four feet 41.4% (229/553) of the time (Wylie et al., 2013). Conversely, it was argued by other experts that clinical cases often have one forefoot that appears to be affected more severely than the other, and this may present as difficulty in lifting the contralateral limb, with the more affected limb appearing relatively easy to lift (in comparison). The “modified Obel” method requires that each forelimb must be easy to lift (forelimb lift prompt and willingly maintained in each forelimb) to score a 0 for this criterion. When conducting the examination, if the forelimb lift is not prompt and willingly maintained in either limb, a score of either 1 or 2 must be allocated to the animal for this criterion. Hence, even in cases where one forelimb is more painful than the other, the overall “modified Obel” score is not affected as a negative score (of 0) is only to be given if the foot lift is prompt and willingly maintained in each forelimb.

Difficulty turning in a circle is another common laminitis sign, occurring in more than 70% of laminitic horses (Wylie et al., 2013). The “modified Obel” method does examine gait at the circle, but it is not required to be assessed if the horse shows severe difficulty walking (Table 2). This is important for welfare reasons and to avoid forced locomotion exacerbating lamellar pathology. Limiting movement is a key treatment strategy preventing further damage to already weakened lamellar tissue (van Eps, 2010). Similarly, for the same reason, trotting is not examined, in line with veterinary practitioners avoiding trotting laminitic horses in over 50% of cases (Wylie et al., 2013).

A bounding digital pulse has been reported to be the most common clinical sign of laminitis (Wylie et al., 2013). Indeed, it was present in 91.2% of laminitis cases, as compared to 23.1% of non-laminitic lameness cases (Wylie et al., 2016). Forelimb digital pulse palpation is an easy task for an equine practitioner to perform, although one limitation of this study is that no data were obtained on the repeatability and reproducibility of this criterion.

Criteria that were included in the prototype method, but later excluded from the final method, were a reluctance to move and the response to hoof testing. The movement criterion was simply absorbed into the description of “gait at the walk” to better describe the laminitis gait. Hoof testing gave a positive response in too many ponies without clinical laminitis (48%) and so was removed (Meier et al., 2018a), despite the fact that the inter-observer agreement for this variable was 100%.

Further limitations of this study include the fact that bias could have occurred if the veterinarians were aware that one of the methods that they were using to assess laminitis cases was indeed the Obel method. Although no titles were given to the methods, and the cases were assessed using the methods in alternating sequence, many veterinarians are familiar with using the Obel method. In comparison to the Obel method, the “modified Obel” method is limited by not prescribing a threshold score at which clinical laminitis occurs. Whereas the Obel method has four well-defined grades, the “modified Obel” method scores individual criteria, and hence assigning a threshold at which laminitis occurs using the “modified Obel” method may result in mis-diagnosis (in the case of non-laminitis lameness cases) and has been avoided in this study. Furthermore, although all efforts were made to make the “modified Obel” method as objective as possible, there will still be a degree of subjectivity between examiners using this system, as witnessed in the intra-observer results where gait at the circle was given the same score 73% of the time, and gait at the walk 81% of the time, as opposed to weight shifting which had a 91% repeatability.

Other limitations include the number of cases explored, the number of veterinarians used to assess them, and the fact that video recordings were used instead of a live assessment for all cases. However, this is not the first study to utilize video analysis for laminitis diagnosis (Menzies-Gow et al., 2010; van Eps & Pollitt, 2009), and others have employed the method as an appropriate research tool for lameness diagnosis (Keegan et al., 1998; Fuller et al., 2006). Additionally, of the 15 videos utilized nine were filmed during a laminitis induction study using a high NSC diet, and hence may not replicate naturally occurring endocrinopathic laminitis entirely. However, it is the intention of the authors to assess the applicability of the “modified Obel” method within veterinary clinical practice using only natural cases in the future. Indeed, feedback was positive in terms of the “modified Obel” method applicability to clinical practice from veterinarians who participated in this study. In addition, the veterinarians who assisted this study by videoing clinical cases also utilized the “modified Obel” method on these cases, and reported that it was easy to understand and use.

The applicability of the “modified Obel” method in terms of owner use for very early detection of laminitis may also be limited, in that many of the signs upon which the criteria are based may not be present in the pre-clinical phase (with the exception of digital pulse). However, the method was not developed for this purpose, and owners and farriers need to be able to recognize symptoms such as pathologic changes in the hoof (dropped soles, abnormal growth rings on the external hoof wall and seedy toe) to detect pre-clinical laminitis (Floyd, 2007; Walsh, 2010).

Conclusions

This study has shown that the “modified Obel” method has excellent repeatability and reproducibility when used to severity score endocrinopathic laminitis cases from both a laminitis induction study and naturally occurring cases. The “modified Obel” method is proposed for use in clinical laminitis trials, as a three-step examination that can allocate a severity score from 0 to 12 using individual clinical criteria. Future development of the “modified Obel” method will include its validation in tracking recovery from laminitis in clinical trials.

Supplemental Information