Foot discomfort is a major concern for patients with rheumatoid arthritis (RA), and nearly 90% of patients complain of foot pain during the course of disease [1,2]. These discomforts can occur during all periods of RA and have a huge impact on the patient’s daily life [3]. Inflammation of the forefoot joints and soft tissue causes forefoot deformity despite proper medical treatment [1,4], and pathologic features around the metatarsophalangeal joint (MTPJ) and interphalangeal joint (IPJ) of the digits, such as hallux valgus, lesser toe deformity and metatarsalgia, are observed in many RA patients [5]. Forefoot joint damage in RA changes pressure distribution patterns due to changes in the anatomical and biomechanical aspect of the MTPJ and increases peak pressure under the forefoot, leading to increased pain during barefoot walking [6]. There have been many advances in pharmacotherapy, including disease-modifying anti-rheumatic drugs (DMARDs) and biological agents, but it is not yet known whether the drug can actually prevent the deformation [7]. Even if pharmacological treatment is performed to alleviate symptoms and maintain function of the feet, 5% to 22% of patients eventually get surgical treatment [8-10].

Several surgical techniques have been described for RA forefoot deformity surgery, and they aim to relieve pain, correct deformity, increase footwear options and restore walking function [11]. They vary by type of procedure on the first MTPJ and lesser toes. Joint preserving surgery can be performed only in cases with mild to moderate deformity without significant arthritic changes of the MTPJ [11,12], thus preserving the MT heads which are important weight-bearing structures. However, for RA patients with severe joint destruction, deformity and bone loss, joint sacrificing procedures are required to relieve symptoms. Regardless of surgical technique, patient’s preoperative expectations have been shown to strongly relate to their ultimate satisfaction [12]. Discussion with physicians about patient’s expectations from surgery is important to prevent or reduce postoperative dissatisfaction. Currently, there does not appear to be any studies on patient’s satisfaction with rheumatoid forefoot deformity (RFD) correction. While there are previous studies on hallux valgus deformity alone, there are no studies that have performed joint sacrificing surgery including lesser toe for RFD. So we designed this study to find out how the surgery affects patients expectations and satisfactions. This study aimed to investigate the clinical expectations, satisfactions and radiologic factors related with overall satisfaction.

A retrospective study of orthopaedic surgery was performed to identify patients who underwent surgical treatment with complete preoperative and minimum 2-year postoperative patient-reported outcome measures. The study was approved by Board of Hanyang University Medical Center (Study number: HYUH 2018-07-015).

Patients who underwent RA forefoot deformity reconstruction, joint sacrificing surgery performed by a single surgeon (I.H.S) at the foot and ankle department at a single institution between January 2000 and April 2016 were enrolled. All the patients were diagnosed as RA, confirmed by rheumatologists according to the American College of Rheumatology RA criteria and the indications for surgery included metatarsalgia associated with severe RFD. All patients had severe hallux valgus deformity with or without arthritic change on the first MTPJ and severe dislocation of more than three lesser MTPJ along with claw toe deformity (Figure 1). Patients with a history of foot or ankle procedures were excluded. All patients provided written informed consent.

Figure 1. Preoperative radiographs of a patient with severe rheumatoid forefoot deformity. HVA, 1-2 IMA are 56.25°, 22.73° (A). Postoperative radiographs of a patient after modified Dwyer procedure. HVA, 1-2 IMA are 27.5°, 23.25° (B). HVA: hallux valgus angle, IMA: intermetatarsal angle.

Surgical treatment was conducted using the modified Dwyer procedure, the joint sacrificing technique. The original Dwyer procedure undergoes arthrodesis of the first MTPJ and proximal IPJ of the lateral four toes, proximal phalanx and metatarsal head resection of the 2nd, 3rd, 4th and 5th rays and their respective tendon interposed to provide a tenodesis effect. The modified version is the same as conventional Dwyer procedure except the IPJ of the lateral four toes and proximal phalanx are preserved if possible and manual reduction of the IPJ along with extensor digitorum longus lengthening and extensor digitorum brevis tenotomy is performed concomitantly.

We performed a perioperative assessment on all patients before surgery and consulted with the rheumatology department to ensure that about those matters and also the preoperative disease activity would not affect perioperative patient’s status. Surgery was performed when preoperative disease activity did not significantly affect the patient’s systemic condition.

Research on expectations and satisfactions was conducted through a questionnaire. The questionnaire asked patients about their preoperative and postoperative concerns and interests, and selected the five most frequently identified items, which patients commonly ask to surgeon before surgery. Questionnaires were administered at preoperative and postoperative outpatient visits or by telephone. A questionnaire of preoperative patient’s expectations and postoperative satisfactions for greater toe and lesser toes were given, based on a 5-digit-scale (5: very high, 4: high, 3: average, 2: low, 1: very low) of common factors for patient’s expectations and satisfactions of forefoot surgery, including repression of relapse in deformity (D), pain reduction (P), improvement in shoe wearing (S), barefoot activity (B) and appearance (A). Overall satisfaction, which also based on a 5-digit-scale (5: very high, 4: high, 3: average, 2: low, 1: very low) and pain were assessed using the visual analog scale (VAS).

Radiologic evaluation was performed using the standing anterior-posterior and lateral view of the foot. For evaluation of forefoot deformity, the hallux valgus angle (HVA) and 1, 2 intermetatarsal angle (IMA) were measured preoperatively and postoperatively. During routine outpatient visits, the degree of fusion of the first MTPJ, maintenance of metatarsal parabola and other anatomical alignment are measured.

Frequency analysis was performed on preoperative patient’s expectations, postoperative satisfactions. To compare changes in of VAS scores and radiologic parameters of HVA and IMA, preoperative and postoperative values were compared using paired t-tests. Pearson’s correlation analysis and multiple linear regression analysis were used to evaluate the relationships between clinical and radiologic factors, including VAS, IMA, HVA, items about satisfaction, preoperative Disease Activity Score 28-erythrocyte sedimentation rate (DAS-28-ESR) score and overall satisfaction. Preoperative the DAS-28-ESR score based on 28 joints (DAS-28) is calculated from four components: number of tender joints, number of swollen joints, VAS score of the patient’s global health, and ESR. DAS-28-ESR score was measured to validate RA disease activity.

Statistical analysis was performed using the SPSS Statistics software (version 24.0; IBM Corp., Armonk, NY, USA). Statistical significance was set at p<0.05.

Thirty-five patients (40 feet) were retrospectively reviewed. One patient was male, and 34 patients were female. The mean age of patients at the time of surgery was 55.1 years (30~72 years), and the mean follow-up period was 9.2 years (2.8 to 20.5 years) (Table 1). Postoperatively, there were no neurologic complications, vascular injury, breakage of instrument or fixation failure. For the greater toe, average patient’s expectations (D, P, S, B, and A) were 4.2, 4.1, 3.0, 2.5, 2.7, respectively and satisfactions were 4.2, 4.0, 3.4, 3.5, 3.3, respectively. For the lesser toes, patient’s expectations (D, P, S, B and A) were 3.9, 4.1, 3.4, 3.0, 2.8, respectively and satisfactions were 3.4, 4.0, 3.4, 3.6, 2.9, respectively (Table 2). D and P were the highest patient’s expectations, followed by S, B and A in descending order. P and D had the highest satisfaction, followed by B, S and A in descending order. The number of cases in each of the 5-digit-scale for each patient’s expectations and satisfactions for items were counted (Table 3).

Table 1 . Demographic data, radiologic measurements and clinical outcomes of patients.

	All cases (n=40)
Age (yr)	55.1±8.2
Sex (male/female)	1/39
BMI (kg/m2)	22.2±2.7
Preoperative
HVA (°)	47.7±9.7
IMA (°)	16.3±4.5
VAS	7.3±2.1
DAS-28	4.56±1.53
Postoperative
HVA (°)	14.0±4.2
IMA (°)	9.7±3.6
VAS	2.2±1.8
Overall satisfaction	4.0±0.8
ΔHVA (°)	37.47±10.51
ΔIMA (°)	5.67±3.24
ΔVAS	5.80±1.94

Values are presented as mean±standard deviation or number only. BMI: body mass index, VAS: visual analog scale, HVA: hallux valgus angle, IMA: 1, 2 intermetatarsal angle, ΔHVA: difference between preoperative HVA and postoperative HVA, ΔIMA: difference between preoperative IMA and postoperative IMA, ΔVAS: difference between preoperative VAS and postoperative VAS, DAS-28: Disease Activity Score 28-erythrocyte sedimentation rate score..

Table 2 . Five-digit-scale of patient’s expectation and satisfaction for rheumatoid forefoot surgery.

Item	Greater toe			Lesser toes
	Expectation	Satisfaction		Expectation	Satisfaction
Repression of relapse in deformity	4.2 (3 to 5)	4.2 (2 to 5)		3.9 (2 to 5)	3.4 (2 to 5)
Progression reduction in pain	4.1 (2 to 5)	4.0 (1 to 5)		4.1 (2 to 5)	4.0 (2 to 5)
Improvement in shoe wearing	3.0 (1 to 5)	3.4 (2 to 5)		3.4 (1 to 5)	3.4 (1 to 5)
Improvement in barefoot activity	2.5 (1 to 5)	3.5 (2 to 5)		3.0 (1 to 5)	3.6 (1 to 5)
Improvement in appearance	2.7 (1 to 5)	3.3 (2 to 5)		2.8 (1 to 5)	2.9 (1 to 5)

5-digit-scale: 5, very high; 4, high; 3, average; 2, low; 1, very low..

Table 3 . Frequency analysis of patient’s expectation and satisfaction.

Item	No. of expectation/satisfactionfor greater toe						No. of expectation/satisfactionfor lesser toe
	Very high (5)	High (4)	Average (3)	Low (2)	Very low (1)		Very high (5)	High (4)	Average (3)	Low (2)	Very low (1)
Questionnaire D	24/23	0/1	16/15	0/1	0/0		14/11	9/10	15/8	2/6	0/5
Questionnaire P	20/22	11/2	3/12	6/3	0/1		22/20	3/1	13/18	2/1	0/0
Questionnaire S	15/11	2/3	2/17	10/9	11/0		7/7	10/9	14/19	8/4	1/1
Questionnaire B	3/9	11/5	2/22	11/4	13/0		7/13	5/1	12/24	12/2	4/0
Questionnaire A	7/7	3/3	10/23	10/7	10/0		2/2	2/0	25/29	9/8	2/1

D: repression of relapse in deformity, P: reduction in pain, S: shoe wearing, B: barefoot activity, A: appearance..

The mean overall satisfaction score postoperatively was 3.95 (2 to 5), that meaning most patients scored average to high for the questionnaire. The number of cases in each of the 5-digit-scale were counted (Table 4). The mean reduction of overall VAS was 5.1 (7.3 to 2.2; p<0.001) (Table 1). Radiologic results showed a significant decrease of HVA after surgery, 47.7±9.7 (30 to 70) to 14.0±4.2 (6 to 21; p<0.001), and IMA, 16.3±4.5 (10 to 25) to 9.7±3.6 (4 to 17; p<0.001) (Table 1). During routine outpatient visits, 1st MTPJ were all united in the greater toe, and 2nd to 5th MTP joints all were resected properly without any significant remnants in the lesser toes. In two cases, postoperative wound dehiscence occurred, which was treated with periodic wound dressings in outpatient. Within two weeks, their wounds were healed clearly without further surgical intervention. No other complications were showed.

Table 4 . Frequency analysis of overall satisfaction.

Overall satisfaction (grade)	Very high (5)	High (4)	Average (3)	Low (2)	Very low (1)
No. of cases	11	17	11	1	0

The mean DAS-28-ESR score was 4.56 (3.09 to 6.09) (Table 1). The average time from RA diagnosis to surgery was 13 years and 9 months. In medications taken for RA before surgery, 28 patients were taking NSAIDs, 30 patients were taking DMARDs, and 18 patients were taking corticosteroid.

In the Pearson correlation analysis, All clinical factors and radiologic factors except preoperative DAS-28-ESR score were positively correlated with overall postoperative satisfaction to moderate and low degrees. Satisfactions with P (p=0.002) and B (p=0.023) for the greater toe, P (p=0.035) and B (p=0.002) for the lesser toes, and the difference between preoperative and postoperative VAS (ΔVAS) (p=0.003) were significantly related to overall satisfaction (Table 5).

Table 5 . Pearson correlation analysis between clinical factors (VAS, IMA, HVA and items for satisfaction) and overall satisfaction.

	Pearson coefficient	p-value
Repression of relapse in deformity in greater toe	0.021	0.897
Progression reduction in pain in greater toe	0.420	0.002
Improvement in shoe wearing in greater toe	0.246	0.127
Improvement in barefoot activity in greater toe	0.359	0.023
Improvement in appearance in greater toe	0.280	0.080
Repression of relapse in deformity in lesser toes	0.224	0.164
Progression reduction in pain in lesser toes	0.334	0.035
Improvement in shoe wearing in lesser toes	0.251	0.118
Improvement in barefoot activity in lesser toes	0.478	0.002
Improvement in appearance in lesser toes	0.256	0.110
ΔHVA	0.072	0.659
ΔIMA	0.003	0.984
ΔVAS	0.459	0.003
Preoperative DAS-28	–0.092	0.286

HVA: hallux valgus angle, IMA: 1, 2 intermetatarsal angle, VAS: visual analog scale, ΔHVA: difference between preoperative HVA and postoperative HVA, ΔIMA: difference between preoperative IMA and postoperative IMA, ΔVAS: difference between preoperative VAS and postoperative VAS, DAS-28: Disease Activity Score 28-erythrocyte sedimentation rate score. Statistically significant (p<0.05)..

Factors correlating significantly with overall satisfaction, were satisfactions with P and B in both the greater and lesser toes, and reduction amounts of VAS scores following surgery. Additionally, we performed multiple linear analyses to identify the independent factors associated with overall satisfaction. Satisfaction with P for the greater toe was significantly associated to overall satisfaction (p=0.042; Table 6).

Table 6 . Multilinear regression analysis between clinical factors and overall satisfaction.

	Unstandardized coefficients				Standardized coefficients
	B	SE	β		t	p-value
Constant	2.061	0.621			3.320	0.003
Progression reduction in pain in greater toe	0.238	0.113	0.347		2.113	0.042
Improvement in barefoot activity in greater toe	–0.070	0.215	–0.082		–0.325	0.747
Progression reduction in pain in lesser toes	–0.059	0.135	–0.075		–0.436	0.666
Improvement in barefoot activity in lesser toes	0.191	0.149	0.235		1.279	0.210
ΔVAS	0.076	0.056	0.230		1.357	0.184

VAS: visual analog scale, ΔVAS: difference between preoperative VAS and postoperative VAS. Statistically significant (p<0.05)..

It is overt that the foot, in particular the forefoot, is a major part in the surgery of inflammatory joint disease. Synovitis of the MTPJ of foot is often the common findings of RA and results in forefoot pain. It is often the initial symptom of RA. It is reported that within the first three years of RA, approximately 65% of the patients have MTPJ involvement [1,13,14]. Furthermore, It is estimated, with disease progression, two thirds of patients have MTPJ pain of hallux, and subluxation or dislocation of lesser MTP joints. Pain, deformity, and dysfunction of the forefoot, not responding to conservative treatment, eventually causes these patients to undergo surgical treatment [8,9].

It is well known that there is a difference between the patient’s expectations undergoing surgery and the surgical goals of surgeons in the orthopedic field [15-17]. Before surgery, physicians should discuss the patient expected results of the surgery with the patient and understand the patient’s expectations as accurately as possible. If patients and surgeons have similar expectations or explain the difference between postoperative satisfactions and preoperative patient’s expectations, patient satisfaction may be higher than if they have different expectations or not explain. A previous study on preoperative patient’s expectations in hallux valgus patients without RA, and they reported that the patients most expected improvement in foot function, followed by relief of pain at the bunion site, comfortable shoe wearing, and decreased pain at the lesser toes [18].

In our study, patients had a mean DAS-28-ESR score of 4.58, indicating that they had, on average, moderate RA activity. There are several studies about relationship between disease activity and surgical treatment in patients with RA. However, most of these study have been conducted in patients who have undergone large joint surgery or artificial joint replacement surgery. On the contrary, our procedure was a small joint surgery and did not constitute an artificial joint replacement. There are no study about high RA disease activity have been associated with postoperative complications or infections in foot and ankle surgery. We beilieved that patient’s symptoms were the result of deformity of the forefoot caused by the chronic course of the disease rather than an inflammatory response. So we did not use the preoperative DAS-28-ESR score as a criteria for making a surgical decision [19].

We analyzed the correlation between preoperative DAS-28-ESR score and patients overall satisfaction and found no significant correlation. We believe that this would be resulted from their symptoms were the result of deformity of the forefoot caused by the chronic course of the disease rather than an inflammatory response.

For the great toe, repression of relapse in deformity was the paients’ most expected factor of RA forefoot deformity surgery, followed by pain, comfortable shoe wearing, barefoot activity and appearance. For lesser toes, pain was the most expected factor, and followed by repression of relapse in deformity, comfortable shoe wearing, barefoot activity and appearance (Table 2). This showed that the factors that patient’s most expected improvement for both the greater toe and lesser toes were pain and repression of relapse in deformity.

Traditionally, joint sacrificing procedures such as 1st MTPJ fusion or resection arthroplasty in 1st MTPJ and resection arthroplasty of the lesser toes are performed for patients with severe RFD. However, with recent advances in RA drug medication, many surgeons have conducted joint preserving surgeries in patients with minimal erosion of the MTP joint and have reported satisfactory results [20]. In the greater toe, 50% to 95% good correction with osteotomy has been reported in rheumatoid hallux valgus [12,21]. In the lesser toes, the Weil or other metatarsal head preserving osteotomies may have the advantage of preserving the plantar attachment and MTPJ function to bear weight [22]. Niki et al. [11] evaluated a combination of joint preserving procedure for rheumatoid forefoot deformities in 30 patients. They reported no cases of nonunion, deformity recurrence, or callosity. However, joint preserving procedures could not be performed to correct severe deformities as in our study.

In contrast, in the greater toe, many procedures are performed for severe rheumatoid forefoot deformities, such as resection arthroplasty or arthrodesis as joint sacrificing procedures. Torikai et al. [23] found that significant improvement in the HVA could not only be achieved through resection but also with arthrodesis. In addition, Horita et al. [24] reported that arthrodesis of the MTP joint should be indicated for severe hallux valgus with an HVA >50°. In the lesser toes, resection arthroplasty has been commonly used in patients with RA and is essentially unchanged from the original description by Hoffman in 1911. Dai et al. [25] reported that resection arthroplasty of the lesser metatarsals, combined with arthrodesis of the first MTP joint, achieved significant improvements in pain relief, deformity correction, and footwear comfort. In our study, the joint sacrificing, Modified Dwyer procedures were conducted for all patients with severe rheumatoid forefoot deformities and overall satisfaction showed improved clinical and radiological results.

In the Pearson correlation analysis, statistically significant clinical factors were satisfactions of P and B for the greater toe, P and B for the lesser toes, and ΔVAS. Above these, satisfaction of P for the greater toe significantly affected overall satisfaction in multiple linear regression analysis. We believe that all items about satisfaction were positively correlated with overall postoperative satisfaction to a moderate or low degree. Although postoperative overall satisfaction is influenced by a variety of factors, there are factors correlated significantly with overall satisfaction. Satisfactions with P and B in both the greater and lesser toes, and reduction in VAS scores following surgery is significantly correlated.

Clinical factors, particularly satisfactions with S and A for the greater and lesser toes, were not significantly related to overall satisfaction. We believe that the reason was the joint sacrificing procedure have disadvantages in aspect of S and A for the greater and lesser toes.

Piqué-Vidal and Vila [26] studied severity of hallux valgus deformities according to angular measurements in 301 radiographs. In this study, preoperative radiologic parameters of all patients showed almost severe deformities. And postoperative radiologic parameters of all patients were mostly corrected within acceptable range. Nonunion, malunion, or other complications on radiologic examinations may affect clinical outcomes. However, in this study, 1st MTP joints were all united in the greater toe, and 2nd-5th MTP joints all were resected without any significant remnants in the lesser toe, and no complications were found. So we believe that those consistent good results in each patient could not make any significant differences in overall satisfaction.

The limitations of this study are as follows. First, this was a retrospective study. Second, the number of patients who underwent the surgery as rather small. However, RA generally has a low prevalence of approximately 1%, and it is difficult to recruit severe RA patients who underwent surgery by the joint sacrificing surgery with long term follow-up. Nevertheless, this is a long term study about the joint sacrificing procedure. This study analyzes not only the patient’s expectations of preoperative patients but also satisfactions after surgery. Our results show that joint sacrificing procedure in RA is one of the surgical methods that can be tailored to acceptable patient satisfaction. These results will be useful for orthopedic surgeons and rheumatologists when discussing surgical treatment with patients for severe RFD.

Although forefoot reconstruction with a joint sacrificing procedure is non-physiological, it could be a good surgical option for severe RFD. Because clinical outcomes and patients overall satisfaction have shown to be improved in most cases. When comparing patients’ expectations and postoperative satisfactions with the joint sacrificing procedure, both the greater and lesser toes showed better satisfaction on items, ‘reduction in pain’ and ‘barefoot activity’ than on other items, ‘shoe wearing' and ‘appearance’.

Since many clinical factors could affect the overall satisfaction and patient’s expectations of individual patient were diverse, it seems to be important to inform patients preoperatively with a detailed explanation of what could and could not be improved much after surgery.

None.

None.

No potential conflict of interest relevant to this article was reported.

Conceptualization: I.H.S. Data curation: S.J.K., Y.W.G. Methodology: I.H.S. Supervision: I.H.S. Writing – original draft: S.J.K., Y.W.G. Writing – review & editing: I.H.S., S.J.K., Y.W.G.