Association Between Risk of Relapse and Type of Surgical Procedure for Raynaud’s Disease

Introduction

Raynaud’s disease (RD) is characterized by intermittent vasospasms affecting microvasculature, particularly the arterioles in the extremities [1]. Its classic presentation, often triggered by emotional stress or cold exposure, manifests as a sequence of skin color changes: from white (ischemia) to violet (cyanosis), red (reperfusion), and back to normal, possibly accompanied by pain [2]. Clinically, RD is categorized into primary and secondary forms. Primary RD is idiopathic and generally less severe. In contrast, secondary RD is primarily associated with connective tissue diseases, such as systemic sclerosis and systemic lupus erythematosus, and can present with persistent limb ischemia, including ulceration and gangrene [3]. The pathogenesis of RD remains largely unknown, with significant involvement of vessel wall abnormalities, endovascular factor release, and autonomic regulation. Unfortunately, no curative treatment for RD exists. Management strategies involve smoking cessation, minimizing cold exposure, and emotional stress avoidance to reduce the frequency of attacks. Conservative therapies include the use of calcium channel blockers or intravenous iloprost for patients unresponsive to standard treatments, although the efficacy of these conservative approaches remains suboptimal [4]. Surgical interventions may be considered for patients who do not respond to conservative treatment or experience severe manifestations leading to limb ischemia, ulceration, or gangrene.

The primary surgical approach for RD is thoracoscopic thoracic sympathectomy [5]. However, this method requires general anesthesia and endotracheal intubation and carries inherent risks, substantial costs, and the potential for severe complications and high recurrence rates. Therefore, there is a need for a cost-effective and minimally invasive procedure that ensures safety and efficacy. With advances in minimally invasive surgery, endoscopic thoracic sympathectomy has emerged as the preferred choice. To further minimize patient trauma, techniques such as computed tomography (CT)-guided chemical neurolysis and radiofrequency denervation have been introduced [6]. In this study, we conduct a follow-up investigation to evaluate the effectiveness and prognosis of these distinct surgical approaches for Raynaud’s disease. Our analysis also explores the factors influencing recurrence of postoperative RD.

Material and Methods

PARTICIPANTS:

This study was approved by the Medical Ethics Committee of our hospital. The data of 134 patients who received percutaneous transcutaneous radiofrequency thermocoagulation from January 2012 to December 2023 in our hospital were evaluated.

INCLUSION AND EXCLUSION CRITERIA:

Inclusion criteria encompassed patients meeting the following conditions [7]: (1) a confirmed diagnosis of Raynaud’s disease; (2) a lack of response to conservative therapies; and (3) individuals who underwent either CT-guided percutaneous sympathomimetic chain chemical thoracic sympathectomy (CTS) or CT-guided percutaneous transthoracic sympathomimetic chain radiofrequency thermocoagulation (RF-TC).

Exclusion criteria were: (1) patients undergoing alternative procedures (CTS or RF-TC) during the 1-year follow-up period; (2) individuals unable to complete clinical assessments due to exceptional circumstances; and (3) patients with incomplete or unavailable contact information.

OBSERVATION INDICATORS:

This study incorporated several observational parameters. Visual Analog Scale (VAS) scores were utilized at distinct postoperative time points, graded on a scale from 0 to 10, where 0 signified the absence of pain and 10 represented severe pain. The Pittsburgh Sleep Quality Index (PSQI) quantified sleep quality using scores ranging from 0 to 21, with lower scores indicative of improved sleep. Disabilities of the Arm, Shoulder, and Hand (DASH) was employed to evaluate upper extremity function, with scores spanning from 0 to 100, where a score of 0 denoted normal upper-extremity function and 100 indicated a significant limitation. The dynamic assessment of VAS, PSQI, and DASH scores involved data collection at various surgical time points: preoperative (t0), immediate postoperative (t1), 1 month (t2), 3 months (t3), and 6 months (t4). Participants meeting the follow-up criteria were included in the analysis, with missing data resulting from missed appointments not considered in the corresponding analyses.

CT-GUIDED CHEMICAL THORACIC SYMPATHECTOMY: In the interest of patient safety, the establishment of at least 1 peripheral venous access was a prerequisite for the concurrent monitoring of physiological parameters, including heart rate (HR), blood pressure (BP), pulse oxygen saturation (SpO2), peripheral temperature (T), and perfusion index (PI). The determination of the T4 level and the design of the puncture path were facilitated using a 3-mm interval CT scanner and positioning grid. After meticulous surgical disinfection and anesthesia achieved with 1% lidocaine, a 22-gauge, 10-cm-long needle was cautiously introduced, approaching the intended target (Figures 1A–1C, 2A, 2B). A 3-mL mixture consisting of 2 mL of 1% lidocaine and 1 mL of iopamidol was injected for diagnostic block. The positioning accuracy of the mixture was validated through monitoring changes in T and PI and repeating the CT scan. Subsequently, 5 mL of a mixture containing 4 mL of lidocaine and 1 mL of anhydrous ethanol was injected bilaterally, with intermittent withdrawal intervals. Postoperative monitoring was sustained for an additional 30 minutes, and extension of the observation period was contingent upon the patient’s clinical status.

CT-GUIDED RADIOFREQUENCY THERMOCOAGULATION:

Preoperative preparation mirrored that of CTS, except for using a specialized radiofrequency puncture needle (consisting of a 22-gauge, 15-cm-long needle with a 1-cm active tip) for RF-TC. The procedure involved formulating the appropriate depth and angle for puncturing the thoracic sympathetic region as guided by CT imaging. Following this, a rigorous regimen of surgical disinfection and local anesthesia was meticulously carried out. Patient comfort was carefully attended to as the radiofrequency puncture needle was delicately advanced towards the T4 capitulum costae, as illustrated in Figure 3A–3C. Subsequently, a sensorial test was conducted, employing parameters of 0.5 mA at 50 Hz, and a motorial test using 1.0 mA at 2 Hz. These tests were crucial in ensuring the absence of nerve innervation zones, as indicated by the absence of muscle numbness and twitches. For patients experiencing severe puncture-related pain, opioid analgesics were administered. The radiofrequency procedure was performed at a temperature of 95°C, with each session lasting 300 seconds, and comprehensive monitoring of pertinent indices was meticulously recorded. Postoperatively, the monitoring regimen was adjusted based on the individual patient’s vital signs and clinical condition.

STATISTICAL ANALYSIS:

Data analysis was performed using R (version 4.0.2; 2020-06-22) and SPSS (version 26.0; IBM, Chicago, USA) software, while GraphPad Prism 9 was employed for data visualization. Independent variables were selected following the criterion of 10 events per variable (10 EPVs). Based on prior studies and research expertise, variables such as age, sex, primary disease, disease duration, and surgical procedure were included in the analysis.

Recurrence, defined as the reappearance of symptoms comparable to the preoperative state, was considered positive when observed and negative when no recurrence or caudal amputation occurred during follow-up. The time to recurrence was recorded for all patients experiencing recurrence. In pursuit of a comprehensive understanding of factors influencing surgical outcomes, we conducted one-way Cox regression analysis to explore the relationship between independent variables and prognosis. Statistically significant variables were subsequently incorporated into a multivariate stepwise Cox proportional hazard regression model (forward: LR), while accounting for covariates between independent variables and controlling for potential confounding factors.

We constructed prognostic charts for primary disease (PD) and surgical approach (SA), and T-D and PI-D ROC curves were plotted for multi-indicator co-diagnostics.

Results

BASELINE CHARACTERISTICS:

We conducted an evaluation of data pertaining to 134 patients diagnosed with primary palmar hyperhidrosis between January 2012 and December 2023. After the exclusion of individuals who declined follow-up (n=6) and those who were lost to follow-up (n=18), our final dataset comprised 110 patients for the purpose of statistical analysis. The baseline characteristics of these included patients are presented in Table 1.

Follow-up: Maximum follow-up of 12 months, minimum follow-up of 6 months.

Recurrence: 110 patients were followed up, of whom 10 had surgery on both the hands and the feet and 72 had recurrence.

VAS, PSQI AND DASH SCORES:

The number of cases included in our analysis varied over time, with 110 cases considered for the t0–t3 time period and 107 cases for t4. Our analysis revealed a notable reduction in VAS, PSQI, and DASH scores among patients in the t1–t4 group compared to t0, as detailed in Table 2A and 2B. Patients reported preoperative pain with a median score of 3, which consistently decreased at all postoperative time points, reaching a median score of 1 or 2. In addition, violin plots were generated based on the data from 107 patients who underwent 1 year of postoperative follow-up. These plots demonstrated a significant decrease in VAS, PSQI, and DASH scores at all postoperative time points in comparison to the preoperative period (P < 0.001), as depicted in Figure 4A–4C.

SIX-MONTH FOLLOW-UP: Over the course of a 6-month follow-up period, 69 patients experienced an outcome event. The cumulative incidence curve is shown in Figure 5A. Univariate (UV) and multivariate (MV) Cox proportional hazard regression analyses (Table 3A) revealed that primary disease (UV: HR, 2.396; 95% CI, 1.504–3.815; MV: HR, 1.718; 95% CI, 1.044–2.829), surgical approach (UV: HR, 0.447; 95% CI, 0.279–0.716; MV: HR, 0.454; 95% CI, 0.272–0.76), perfusion index difference (UV: HR, 0.796; 95% CI, 0.707–0.896; MV: HR, 0.870; 95% CI, 0.76–0.994), and peripheral temperature difference (UV: HR, 0.72; 95% CI, 0.608–0.852; MV: HR, 0.755; 95% CI, 0.615–0.928) were independently risk factors influencing the recurrence rate of Raynaud’s disease. Negative associations were observed for the presence of CTS, primary disease, T-D, and PI-D with the primary outcome. The areas under the curve (AUC) for PI-D, T-D, OM, and PD were 0.793, 0.752, 0.686, and 0.680, respectively, as demonstrated in Figure 5B. Figure 5C illustrates the recurrence rate of RD following treatment.

ONE-YEAR FOLLOW-UP: Over the 1-year follow-up period, 72 patients experienced outcome events, as depicted in Figure 6A. Univariate (UV) and multivariate (MV) Cox proportional hazard regression analyses (Table 3B) demonstrated the independence of primary disease (UV: HR, 2.674; 95% CI, 1.668–4.289; MV: HR, 1.881; 95% CI, 1.125–3.145), surgical approach (UV: HR, 0.447; 95% CI, 0.279–0.716; MV: HR, 0.454; 95% CI, 0.272–0.76), perfusion index difference (UV: HR, 0.8; 95% CI, 0.711–0.9; MV: HR, 0.866; 95% CI, 0.759–0.989), and peripheral temperature difference (UV: HR, 1.631; 95% CI, 0.910–2.924; MV: HR, 0.757; 95% CI, 0.618–0.928) as independent risk factors influencing the recurrence rate of Raynaud’s disease. The presence of CTS, primary disease, T-D, and PI-D was inversely related to the primary outcome. The areas under the curve (AUC) for PI-D, T-D, OM, and PD were 0.774, 0.721, 0.683, and 0.668, respectively, as illustrated in Figure 6B. The recurrence rate of RD following treatment is visually represented in Figure 6C.

ADVERSE EVENTS AND SATISFACTION:

Following surgery, 18 out of the 110 patients experienced transient sinus bradycardia lasting 2 to 3 hours, after which they spontaneously reverted to a normal rhythm. In addition, 4 patients had nausea and vomiting, which resolved within 1 day. Patient satisfaction scores, on a scale from 0 to 10 (where 0 signifies very dissatisfied and 10 denotes very satisfied, with ranges of 0–2 indicating dissatisfaction, 3–5 indicating mild satisfaction, 6–8 indicating satisfaction, and 9–10 indicating high satisfaction), yielded an overall patient satisfaction score of 4.27.

Discussion

Raynaud’s disease (RD) is a relatively common disorder, yet its estimated prevalence varies widely, spanning from 1% to 19% [8]. This variability is attributed to divergent definitions of RD and geographic factors. Its predominant clinical manifestation is excessive vasoconstriction triggered by cold exposure, driven by the cold-induced reflex of the sympathetic nervous system and local activation of α2C adrenoceptors within the peripheral circulation [9]. Raynaud’s disease is classified into primary Raynaud’s disease (pRD) and secondary Raynaud’s disease (sRD). The former manifests as an isolated disorder, often characterized by transient paroxysmal vasoconstriction with no discernible vascular structural abnormalities or minimal variations. In contrast, sRD is associated with structural and biochemical alterations, leading to prolonged ischemia and digital necrosis [10]. Notably, in pRD, although antinuclear antibodies are present at low titers, further investigations have identified antinuclear antibodies as independent predictive factors for pRD progression [11]. Research has also highlighted elevated serum levels of nerve growth factor, endothelin-1, and vasoconstrictor angiotensin II in sRD when compared to normal individuals [12,13]. These biochemical alterations contribute to vasoconstriction and vascular structural changes. Genetic studies have revealed a polygenic predisposition to RD [14]. Despite substantial investments in financial and human resources, the intricate pathological mechanisms and the absence of a suitable animal model for RD pose significant challenges in the quest for a cure.

Considering the underlying pathophysiology, the primary focus of RD therapy centers on the dilation of constricted blood vessels. The European League Against Rheumatism (EULAR) recommends calcium channel blockers (CCBs), including oral dihydropyridine and nifedipine, as the first-line pharmacological intervention for RD. Notably, dihydropyridine-class CCBs effectively reduce the frequency and duration of vasospastic episodes, alleviating associated pain and severity, although they may be accompanied by adverse effects such as dizziness, nausea, edema, and headaches [15]. However, it is noteworthy that some studies have suggested that the therapeutic efficacy of CCBs can diminish with prolonged treatment [16]. Of particular concern is a comprehensive review published in 2016, which concluded that CCB drug therapy was largely ineffective in primary RD [17]. Alternative approaches in RD therapy involve the use of angiotensin-converting enzyme inhibitors, like losartan, and angiotensin II receptor antagonists such as quinapril. These medications are often considered in cases where patients experience vasodilatory adverse effects [18,19]. Nevertheless, these drugs can cause adverse effects, including nausea, depression, cough, and, in some instances, severe complications like kidney failure or malignant paroxysmal hypertension, necessitating the discontinuation of this conventional therapy. Furthermore, research has indicated that angiotensin-converting enzyme inhibitors have limited efficacy and do not contribute to the healing of digital ulcers [20]. In cases of severe RD symptoms, such as extremity ulcers and gangrene, newer therapeutic options encompass phosphodiesterase-type 5 inhibitors and prostaglandins [21–23]. However, evidence regarding the efficacy of beraprost in RD treatment remains limited and inconsistent, as shown in a comprehensive review by Stewart [24].

Despite ongoing efforts to develop new pharmacological interventions, an increasing number of patients have shown a preference for invasive treatment modalities. Notably, cervical or thoracic sympathectomy has been employed in cases of refractory RD; however, its inherent invasiveness has limited its widespread adoption. The advent of thoracoscopy has introduced thoracoscopic sympathectomy as an alternative therapeutic approach for RD. In a follow-up study by Thune, immediate efficacy was achieved in 83% of patients after surgery, with 63% experiencing compensatory sweating, 57% reporting dry hand syndrome, and 30% having gustatory sweating. Importantly, all observed adverse effects were either temporary or permanent. The elevated incidence of adverse effects can be attributed to the necessity of complete sympathetic chain resection to ensure the procedure’s efficacy. However, retrospective research has revealed a high recurrence rate of up to 82% within 16 months of follow-up, particularly among those with underlying autoimmune diseases [25]. Data from these investigations emphasize that compensatory sweating is the most common and bothersome complication, widely ranging from 3% to 98% [26,27]. Remarkably, only 8% of patients undergoing thoracoscopic sympathectomy achieved an excellent long-term effect, while 50% subjectively reported a beneficial outcome from the procedure [28]. Hence, there is a pressing need for innovative operations with reduced invasiveness and lower recurrence rates in the management of RD.

In our medical institution, sequential chemical thoracic sympathectomy (CTS) and radiofrequency thermocoagulation (RF-TC) has been employed in patients with recalcitrant, inappropriate, or unresponsive RD. A comprehensive follow-up survey involving 110 RD patients found that 10 patients underwent surgeries on both the hands and the feet, with 59 undergoing CTS. Forty-four of these CTS patients experienced recurrence at the 6-month follow-up, and this recurrence increased to 47 at 12-month follow-up, signifying a substantial recurrence rate, accompanied by postoperative adverse effects. Notably, the observed postoperative adverse effects included Horner’s syndrome, which manifested in 57% of patients in the immediate postoperative period, and compensatory hyperhidrosis, observed in 68% of patients. These adverse effects predominantly stem from the unpredictable and challenging-to-control distribution of anhydrous ethanol injection, often resulting in extended burning pain or varying degrees of demyelination. In a proactive measure to circumvent these complications, we introduced RF-TC as an alternative and selective treatment for RD patients. After the 6-month follow-up, 61 patients underwent RF-TC surgery, with 25 experiencing recurrence. This trend continued at the 12-month follow-up, with 54 patients receiving RF treatment and 25 having recurrence. It is worth noting that the recurrence rate with RF-TC was markedly lower than that observed with CTS. Moreover, the incidence of postoperative complications, such as compensatory hyperhidrosis, decreased to 3.6%, and Horner’s syndrome incidence was reduced to 1%. These findings are consistent with earlier research, highlighting the notably lower occurrence of neuralgia and motor or sensory nerve injuries associated with RF-TC compared to CTS, reaffirming the alignment of our results with prior studies [29]. This is because CTS can destroy sympathetic nerves through the diffusion of alcohol, whereas in RF-TC the placement of the RF needle must be more precise, and the closer the tip of the needle is to the sympathetic chain, the more complete the destruction of the sympathetic nerves. This also explains why RF-TC has fewer complications than CTS.

In our investigation, we incorporated the pulsatile index (PI), denoting the ratio of pulsatile to non-pulsatile signals, into our set of observational parameters [30]. Prior research endeavors have utilized PI as a robust and sensitive metric for assessing the success of peripheral nerve blocks [31,32]. Our data unequivocally confirm a significant immediate increase in PI following radiofrequency thermocoagulation (RF-TC), signifying vasodilation induced by sympathectomy. Encouragingly, our subsequent statistical analyses have firmly established an inverse relationship between changes in PI and the recurrence rate. Concurrently, we observed consistent correlations between fluctuations in skin temperature during surgery and these outcomes. To elucidate, individuals with RD who exhibited more substantial increases in PI and skin temperature postoperatively tended to achieve more enduring therapeutic effects. As previously reported in the literature, the management of RD proves to be a more formidable challenge in patients with underlying comorbidities. Consistent with these earlier findings, our current study reveals a heightened susceptibility to relapse among patients with underlying diseases.

To conduct a comprehensive evaluation of therapeutic efficacy, we incorporated the Visual Analog Scale (VAS), Pittsburgh Sleep Quality Index (PSQI), and Disabilities of the Arm, Shoulder, and Hand (DASH) as subjective measures. Encouragingly, for most patients, VAS and DASH scores exhibited significant reductions during the early follow-up phase. However, it is important to acknowledge that both evaluation indices demonstrated a mild rebound in later stages. In contrast, the PSQI score only exhibited a slight reduction.

Notably, the CT-guided percutaneous sympathectomy clearly heightened the patients’ tolerance for fine motor abilities and enhanced their overall quality of life. Recurrence of RD is associated with several factors, including incomplete surgery, reinnervation of sympathetic pathways, elevated sensitivity of noradrenergic receptors, and the presence of connective tissue disease.

Conclusions

LIMITATIONS:

This was a non-randomized, single-center, retrospective study and could not exclude outliers. Additionally, the small sample size used does not allow further segmentation or validation of the findings.