Analysis of clinical features of lupus mesenteric vasculitis

Abstract

This study aimed to identify the clinical characteristics of systemic lupus erythematosus (SLE) patients with complicated with lupus mesenteric vasculitis (LMV), and to determine the impacts of the first-visit department, the number and duration of digestive system symptoms on disease activity, and imaging features of LMV patients. We expected that our study would keep clinicians across specialties vigilant for SLE-LMV, enabling early diagnosis and prompt treatment, thereby improving prognosis and reducing mortality from this potentially fatal condition. Clinical data of 174 patients with LMV admitted to the First Affiliated Hospital of Zhengzhou University from January 2015 to December 2023 were collected and analyzed. The baseline clinical characteristics and laboratory data were analyzed and compared by the first-admission department: either the Rheumatology Department or other departments. In addition, the impact of the numbers and durations of various digestive system symptoms in these patients, as well as laboratory results, disease activity, and imaging findings, was analyzed. The independent-samples t-test and Mann-Whitney U-test were used to compare differences in continuous variables. In contrast, the Kruskal-Wallis test, chi-square test, or Fisher’s exact test was used to compare differences in disease activity-related indicators, imaging features, and treatment regimens. A total of 174 patients with an initial diagnosis of LMV were included in this study. Among these patients, 87 (50.0%) had a prior SLE diagnosis, 76 (43.7%) were first diagnosed in the Rheumatology Department, and 36 (20.7%) and 31 (17.8%) were first diagnosed in the Departments of Gastroenterology and Emergency Medicine, respectively. All patients had mild to severe digestive symptoms, including abdominal pain, diarrhea, nausea, vomiting, and abdominal distension. Multiple digestive symptoms occurred in 151 (86.8%) patients. The patients were divided into two groups based on whether they were first diagnosed in the rheumatology department or other departments. The results showed that patients who were not initially diagnosed in the rheumatology department had longer hospital stays and lower disease activity. The patients were further categorized into groups based on the number and time of various digestive symptoms. Compared with those with fewer flares, the proportions of leukopenia, hypocomplementemia, and abdominal effusion were higher in those with frequent LMV recurrences. The white blood cell (WBC) and neutrophil counts in patients with abdominal pain of less than 7 days were higher than those with abdominal pain of more than 7 days. However, statistical analysis did not reveal significant differences among other inflammatory markers and the indices of disease activity. In terms of imaging features obtained by contrast-enhanced CT, a statistical difference was detected in patients with abdominal pain over 7 days compared to those with less than 7 days with respect to thickening of both small and large bowel walls. Due to atypical clinical manifestations of SLE patients with concomitant LMV, only approximately 50% patients were first diagnosed at the Rheumatology Department, with the rest being first diagnosed in other departments, which resulted in delayed diagnosis and treatments, prolonged hospital stay, and potentially, poor prognosis. Evaluation of the impacts of the number and duration of digestive symptom flares on disease activity and imaging features demonstrated that patients with multiple symptoms had more severe radiological findings. Therefore, clinicians, particularly rheumatologists, should keep vigilant when patients have multiple digestive system symptoms and frequent recurrence, as early diagnosis of LMV with prompt and aggressive treatments may improve the prognosis of these patients.

Introduction

Systemic lupus erythematosus (SLE) is a highly heterogeneous autoimmune disease. Loss of immune self-tolerance leads to robust autoantibody production and the formation of immune complexes. The subsequent release of a wide array of inflammatory factors causes damage to virtually all systems/organs [1, 2]. Although gastrointestinal (GI) symptoms are common in patients with SLE, more than half are attributable to adverse drug reactions and infections, whereas GI involvement per se is rare. SLE-induced GI manifestations include protein-losing enteropathy, hepatitis, pancreatitis, and pseudo-intestinal obstruction [3–5]. Lupus mesenteric vasculitis (LMV), is one of the most severe GI complications in patients with SLE. Because of its protean manifestation, LMV is frequently considered a drug-induced adverse reaction, leading to delayed diagnosis. Without prompt intervention, LMV may become a fatal condition because of GI bleeding and perforation [6, 7]. At present, the diagnosis of LMV is based on clinical features, imaging findings, and histopathological examinations. Regarding imaging techniques, abdominal computed tomography (CT) is widely accepted as the gold standard for diagnosis. Early identification and prompt treatment are crucial for improving prognosis and reducing mortality [6, 8].

The majority of current investigations on LMV consist of case reports or small-sample clinical series. In particular, there is a marked scarcity of research on the clinical characteristics associated with disease activity and the imaging features of newly onset LMV [9–11]. Most patients do not present to the rheumatology department for their initial consultation because GI symptoms are nonspecific [3]. As such, the association between initial consultation departments and disease activity, treatment selection, prognosis, and mortality remains unclear. In addition, no systematic studies have been conducted to examine the effects of the number and duration of GI symptom flares on the clinical course, disease activity, laboratory results, and changes in imaging features in patients. The existence of this research gap poses a significant challenge for clinicians, hindering their understanding of the broader context of LMV and potentially leading to delayed diagnosis and treatment.

In this study, we first retrospectively reviewed the clinical and laboratory characteristics of LMV patients and compared the impacts of different first-visit departments on disease progression. Meanwhile, based on the number and duration of patients’ GI symptom flares, we compared the differences in their clinical features, laboratory test results, imaging findings, and disease activity. Our objective was to facilitate early diagnosis and timely intervention for LMV among clinicians across disciplines, thereby improving treatment outcomes and prognosis.

Materials and methods

Study Population

A total of 174 patients with SLE complicated with LMV, who were first diagnosed in our hospital from January 2015 to December 2023, were enrolled in this study. All patients met the revised SLE classification criteria issued by the American College of Rheumatology (ACR) in 2009, and none had a history of glucocorticoid or cytotoxic drug use [12, 13]. All LMV patients met the following diagnostic criteria: 1) clinical evidence of multifocal intestinal involvement (circumferential segmental involvement of the small and large intestines with skip lesions), involvement of multiple vascular regions (affecting multiple anatomical areas without being confined to a single vascular supply area), duodenal ischemic changes (ischemic changes detected by CT or segmental edema and discontinuous ulcers detected by endoscopy, intestinal wall thickening, and clinical improvement after treatment with intravenous steroids or immunosuppressants, 2) Abdominal CT confirmed the presence of at least 3 of the following signs: intestinal wall thickening, target sign, intestinal segment dilation, mesenteric vessel engorgement, and increased mesenteric fat density [14]. Intestinal wall thickening was defined as a wall thickness of at least 3 mm in the maximally distended region of the intestine, and 5 mm for the gastric wall [15, 16]. Intestinal segment dilation was defined as a small intestine diameter greater than 2.5 cm and a large intestine diameter greater than 8.0 cm [17, 18]. Patients were excluded if they had drug-induced GI discomfort, primary GI diseases, GI manifestations that could be explained by SLE involvement of other organs, atypical GI manifestations without objective evidence, or comorbidities of other rheumatic and immunological diseases, such as antiphospholipid antibody syndrome.

Data Collection

Clinical and demographic information was collected from our medical records of hospitalized patients, including gender, age, duration of SLE, duration of abdominal pain, and involvement of important organs. Laboratory test results were also collected, including routine blood tests, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), complement C3, complement C4, antinuclear antibodies (ANA), anti-Sm antibodies, anti-ribosomal antibodies, anti-Ro antibodies, anti-cardiolipin antibodies, and anti-β2-glycoprotein antibodies. The Neutrophil-to-Lymphocyte Ratio (NLR), Platelet-to-Lymphocyte Ratio (PLR), and Systemic Immune-Inflammation Index (SII) were calculated. Among them, the calculation formula of SII is as follows: SII = peripheral platelet count × absolute neutrophil count × 1/absolute lymphocyte count [19]. Imaging features and patient evaluation: All patients underwent abdominal CT scans upon admission and after abdominal pain resolution. Imaging changes of involved organs (stomach, small intestine, colon, rectum), intestinal wall thickness (mild: 4–5 mm, moderate: 6–8 mm, severe: >8 mm), mesentery, lymph nodes, urinary tract, and ascites were recorded. CT interpretation was performed in a blinded manner by two independent radiologists, and if there was a disagreement, a consensus was reached after thorough discussion [16]. The Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) was assessed by rheumatologists [20, 21]. Treatment regimens were formulated based on treatment guidelines, the availability of medicines, and economic status of patients. The therapeutic drugs and administration regimens of patients were collected, including glucocorticoids and immunosuppressants (cyclophosphamide, mycophenolate mofetil, azathioprine, leflunomide, cyclosporine, tacrolimus, etc.) [1]. Due to differences in disease severity and individual clinical experience of physicians, there were significant variations in the dosages and duration of glucocorticoid therapy. In this study, glucocorticoid dosages were defined as follows: low dose (≤ 40 mg·d⁻¹ prednisone or equivalent), moderate dose (> 40 mg·d⁻¹ and ≤ 100 mg·d⁻¹ prednisone or equivalent), high dose (> 100 mg·d⁻¹ and ≤ 500 mg·d⁻¹ prednisone or equivalent), and pulse therapy (500–1000 mg·d⁻¹ prednisone or equivalent).

Statistical methods

Descriptive statistical analysis was used to summarize the demographic, clinical, and imaging characteristics. Normality tests were performed for continuous variables. For the comparison of continuous variables between two groups, normally distributed continuous variables were expressed as mean (x̄) ± standard deviation (s) and analyzed using an independent samples t-test, while non-normally distributed data were presented as median [M (P25, P75)] and analyzed using the Mann-Whitney U-test. For multiple groups of continuous variables and ordinal outcome variables (e.g., glucocorticoid dosage, severity of intestinal wall thickening), the Kruskal-Wallis test was performed. Categorical data were expressed as numbers and percentages (%), and the chi-square test or Fisher’s exact test was used to compare differences in disease activity-related indicators, imaging features, and treatment regimens. Data analysis was conducted using IBM SPSS Statistics Version 26.0, and a two-tailed P ≤ 0.05 was considered statistically significant. Statistical graphs of patients’ clinical and imaging characteristics were plotted using GraphPad Prism 8 and Excel software.

Results

Analysis of general patient data

A total of 174 patients were enrolled in this study. Among them, 150 (86.2%) were female, and 87 (50.0%) patients had a previous diagnosis of SLE. Only 76 (43.7%) patients’ first-visit department was rheumatology. In contrast, 36 (20.7%) and 31 (17.8%) patients were initially treated in the department of gastroenterology and emergency medicine, respectively. The demographic data and laboratory test results of all patients are shown in Table 1.

Table 1.

Demographic Data and Laboratory Examinations of Patients with LMV

Item	Median [M(P25, P75)]
Length of Hospital Stay(d)	16(10.75, 21.25)
Age(year)	34(22.75, 47.00)
WBC(×109 L-1)	4.92(3.33, 7.92)
Neutrophil(×109 L-1)	3.47(2.34, 5.90)
Lymphocyte(×109 L-1)	0.84(0.54, 1.29)
Platelet(×1012 L-1)	172.50(117.25, 232.25)
Hemoglobin (g·L-1)	108.80(93.00, 123.00)
NLR	4.59(3.00, 6.50)
PLR	203.38(127.08, 335.29)
SII	760.04(393.29, 1375.88)
ESR/(mm·h-1)	20.00(8.00, 53.00)
CRP/(mg·L-1)	5.89(1.90, 20.30)
C3/(g·L-1)	0.47(0.35, 0.58)
C4/(g·L-1)	0.09(0.06, 0.14)

WBC: White Blood Cell Count; NLR: Neutrophil-to-Lymphocyte Ratio; PLR: Platelet-to-Lymphocyte Ratio; SII: Systemic Inflammatory Immune Index; ESR: Erythrocyte Sedimentation Rate; CRP: C-reactive Protein

Further analysis was conducted on the symptoms, disease activity-related evaluation indicators, and imaging findings of all patients (Fig. 1). All patients presented with varying degrees of digestive system symptoms, including abdominal pain, diarrhea, abdominal distension, nausea, and vomiting, among whom 151 (86.8%) had multiple digestive symptoms. Notably, the proportions of patients with disease activity-related symptoms were as follows: fever in 63 (36.2%), alopecia in 26 (14.9%), skin rash in 51 (29.3%), arthralgia and myalgia in 55 (31.6%), oral ulcers in 9 (5.2%), and Raynaud’s phenomenon in 18 (10.3%). For relevant laboratory indicators, the proportions of positive dsDNA antibodies, leukopenia, thrombocytopenia, hypocomplementemia, and 24-hour urinary protein > 0.5 g were 89 (51.1%), 28 (16.1%), 27 (15.5%), 122 (70.1%), and 72 (41.4%), respectively. Regarding imaging diagnostic indicators, the proportions of intestinal wall thickening, target sign, intestinal segment dilation, mesenteric vessel engorgement, and increased mesenteric fat density were 161 (92.5%), 106 (60.9%), 124 (71.3%), 174 (100%), and 155 (89.1%), respectively (Fig. 1).

Fig. 1.

Symptoms, Disease Activity and Imaging Features of Patients with LMV

Glucocorticoids and immunosuppressants were used as first-line regimens for LMV treatment. In this study, 47% of patients received high-dose (25.3%) or pulse-dose (21.8%) glucocorticoids (> 100 mg). Other medications included immunomodulators such as hydroxychloroquine (24.1%), while immunosuppressants were dominated by mycophenolate mofetil (20.1%) (Fig. 2).

Fig. 2.

Analysis of Treatment Regimens in Patients with LMV. (A) Application of Glucocorticoids at Different Doses. (B) Application of Immunosuppressants in Patients

Analysis of first-visit departments in patients

Through analysis, we found that only 76 out of 174 patients were first diagnosed with LMV in the department of rheumatology. Patients were divided into two groups according to whether their first-visit department was rheumatology, and their lengths of hospital stay, laboratory test results, and clinical symptoms were compared. The results showed that: patients who first visited non-rheumatology departments had a longer length of hospital stay (17.50 vs. 11.00, P < 0.001), while their disease activity was lower than that of patients who first visited rheumatology departments (8.00 vs. 10.00, P = 0.002), and the inflammation-related indicators PLR and SII were also lower. In terms of treatment regimen formulation, a relatively higher proportion of patients who first visited non-rheumatology departments did not receive immunosuppressants (17/98), whereas these patients had a higher proportion of receiving pulse-dose glucocorticoids (> 500 mg) during treatment (28.6%).

When comparing the symptoms of both groups, patients who first visited rheumatology department frequently presented with SLE-specific manifestations including alopecia (28.9% vs. 4.1%), skin rashes (40.8% vs. 20.4%), arthralgia and myalgia (55.3% vs. 13.3%), and Raynaud’s phenomenon (21.1% vs. 2.0%). In contrast, patients who first visited non-rheumatology departments more frequently presented with digestive system symptoms such as nausea and vomiting (69.7% vs. 85.7%). (Table 2).

Table 2.

Comparison of demographic data, clinical symptoms and laboratory tests between two groups

	First visit to department of rheumatology(n = 76)	First visit to department of non-rheumatology(n = 98)	P value
Demographic Data
Length of Hospital Stay (d)	11(7.25, 19.75)	17.50(14.00, 23.25)	<0.001
Age (years)	32(22, 44)	34.5(25, 49)	0.299
Female [n (%)]	66(86.8)	84(85.7)	0.831
Previous History of SLE [n (%)]	56(73.7)	31(31.6)	<0.001
History of Recurrent Abdominal Pain [n (%)]	25(32.9)	18(18.4)	0.028
Clinical Symptoms [n (%)]
Fever	30(39.5)	33(33.7)	0.430
Alopecia	22(28.9)	4(4.1)	<0.001
Skin rash	31(40.8)	20(20.4)	0.003
Oral ulcer	5(6.6)	4(4.1)	0.506
Arthralgia and Myalgia	42(55.3)	13(13.3)	<0.001
Raynaud’s Phenomenon	16(21.1)	2(2.0)	<0.001
Abdominal Pain	66(86.8)	89(90.8)	0.404
Diarrhea	31(40.8)	51(52.0)	0.140
Abdominal Distension	41(53.9)	56(57.1)	0.674
Nausea and Vomiting	53(69.7)	84(85.7)	0.011
Laboratory Tests
WBC(×109 L− 1)	4.6(3.3, 7.1)	5.4(3.38, 8.53)	0.279
NC(×109 L− 1)	3.39(2.29, 5.24)	3.52(2.34, 6.57)	0.642
Lymphocyte(×109 L− 1)	0.69(0.50, 1.21)	1.00(0.64,, 1.40)	0.027
Platelet(×1012 L− 1)	184(126, 255)	157(112.75, 217.00)	0.093
Hemoglobin(g·L− 1)	108.8(92.00, 119.00)	108.50(93.75, 125.00)	0.651
NLR	5.09(3.27, 6.52)	4.16(2.83, 6.66)	0.099
PLR	247.76(160.95, 356.06)	168.55(102.88, 285.60)	0.002
SII	966.99(538.37, 1588.74)	655.92(359.92, 1184.43)	0.013
ESR/(mm·h− 1)	17.50(8.00, 55.75)	26.00(8.00, 50.50)	0.588
CRP/(mg·L− 1)	5.55(1.58, 24.67)	5.90(2.16, 20.30)	0.866
C3/(g·L− 1)	0.49(0.35, 0.63)	0.46(0.35, 0.58)	0.700
C4/(g·L− 1)	0.08(0.06, 0.16)	0.10(0.07, 0.14)	0.700
D-dimer	1.67(0.87, 2.64)	2.51(1.11, 4.60)	0.001
Positive Fecal Occult Blood	11(14.5)	31(31.6)	0.009
Disease Activity Assessment and Medication Use
SLEDAI	10.00(7.25, 13.00)	8.00(4.00, 11.00)	0.002
Immunosuppressant Use [n (%)]	69(90.8)	81(82.7)	0.123
Glucocorticoid Dose [n (%)]			0.423
None	0(0)	4(4.1)
Low dose	8(10.5)	6(6.1)
Moderate dose	32(42.1)	42(42.9)
High dose	26(34.2)	18(18.4)
Pulse dose	10(13.2)	28(28.6)
Intravenous Immunoglobulin Use [n (%)]	13(17.1)	26(26.5)	0.139

NLR: Neutrophil-to-Lymphocyte Ratio; PLR: Platelet-to-Lymphocyte Ratio; SII: Systemic Inflammatory Immune Index; ESR: Erythrocyte Sedimentation Rate; CRP: C-reactive Protein; SLEDAI: Systemic Lupus Erythematosus Disease Activity Index

Different digestive system symptoms

Given that a relatively high proportion of patients seek medical attention in the department of gastroenterology or emergency medicine due to digestive system symptoms such as abdominal pain and distension, as well as nausea and vomiting, we will further compare the impact of the number and duration of different digestive system symptoms on patients’ laboratory test results and imaging findings.

Comparative analysis of patients with ≤ 2 symptoms vs. >2 symptoms

Among the enrolled patients, GI symptoms included four types: abdominal pain, diarrhea, abdominal distension, and nausea and vomiting. We recorded the number of GI symptoms for each patient and divided the patients into two groups: low-symptom group (≤ 2 symptoms) and high-symptom group (> 2 symptoms group). Compared with the low-symptom group, the high-symptom group had a lower proportion of patients with a history of prior SLE (42.5% vs. 63.9%, P = 0.007) and a lower proportion of patients who first visited the rheumatology department (35.4% vs. 59.0%, P = 0.003). No statistically significant difference in disease activity was observed between the two groups. However, the proportions of leukopenia (20.4% vs. 8.2%, P = 0.037) and hypocomplementemia (75.2% vs. 60.7%, P = 0.045) in the high-symptom group were higher than those in the low-symptom group. Notably, the low-symptom group had a higher proportion of patients with fever (49.2% vs. 29.2%, P = 0.009), and arthralgia and myalgia (45.9% vs. 23.9%, P = 0.003). No significant differences were detected between two groups regarding alopecia, skin rash, or oral ulcers (Table 3).

Table 3.

Comparison of Clinical Symptoms and Imaging Features Between the Two Groups

	Low-symptom group(n = 61)	High-symptom group(n = 113)	χ2	P value
History of prior SLE [ n(%)]	39(63.9)	48(42.5)	7.295	0.007
First visited the department of rheumatology [n(%)]	36(59.0)	40(35.4)	8.983	0.003
Clinical symptom [ n (%)]
Fever	30(49.2)	33(29.2)	6.845	0.009
Alopecia	13(21.3)	13(11.5)	2.998	0.083
Skin rash	18(29.5)	33(29.2)	0.002	0.966
Oral ulcer	5(8.2)	4(3.5)		0.281
Arthralgia and myalgia	28(45.9)	27(23.9)	8.876	0.003
Raynaud phenomenon	8(13.1)	10(8.8)	0.777	0.378
Laboratory tests [ n (%)]
Leukopenia	5(8.2)	23(20.4)	4.336	0.037
Thrombocytopenia	9(14.8)	18(15.8)	0.042	0.838
Low complement	37(60.7)	85(75.2)	4.011	0.045
Anti-dsDNA positive	26(42.6)	63(55.8)	2.733	0.098
24-hour urine protein > 0.5 g	26(42.6)	46(40.7)	0.060	0.807
Imaging Results [ n (%)]
Pericardial effusion	23(37.7)	27(23.9)	3.690	0.055
Pleural effusion	32(52.5)	69(61.1)	1.204	0.273
Abdominal effusion	33(54.1)	91(80.5)	13.516	<0.001
Multiple serosal effusions	32(52.5)	67(59.3)	0.754	0.385
Bowel wall thickening>5 mm	18(29.5)	28(24.8)	0.456	0.500
Bowel wall thickening>3 mm	51(83.6)	110(97.3)	10.816	0.001
Small bowel wall thickening	30(49.2)	85(75.2)	11.987	0.001
Large bowel wall thickening	32(52.5)	79(69.9)	5.224	0.022
Combined thickening of both small and large bowel walls	11(18.0)	55(48.7)	15.797	<0.001
The severity of bowel wall thickening [n(%)]				<0.001
Mild(3 ~ 5 mm)	35(57.4)	36(31.9)	10.680	0.001
Moderate(6 ~ 8 mm)	8(13.1)	47(41.6)	14.862	<0.001
Sever(>8 mm)	7(11.5)	27(23.9)	3.886	0.049
Target sign	24(39.3)	82(72.6)	18.365	<0.001
Bowel segment dilation	35(57.4)	89(78.8)	8.846	0.003
Mesenteric enhancement	61(100.0)	113(100.0)
Increased mesenteric fat density	55(90.2)	100(88.5)	0.113	0.736
Abdominal lymphadenopathy	33(54.1)	50(44.2)	1.541	0.214
Disease activity
SLEDAI	10.0(7.0, 13.0)	9(6.0, 11.0)		0.100
Immunosuppressants [n(%)]	57(86.4)	79(88.8)	0.203	0.652
Glucocorticoid dosage [n(%)]				0.105
None	3(4.9)	1(0.9)
Low dose	9(14.8)	5(4.4)
Moderate dose	21(34.4)	53(46.9)
High dose	19(31.1)	25(22.1)
Pulse dose	9(14.8)	29(25.7)
Intravenous immunoglobulin [n(%)]	11(18.0)	28(24.8)	1.037	0.309

SLEDAI: Systemic Lupus Erythematosus Disease Activity Index

A comparison of imaging findings showed that the high-symptom group had a significantly higher proportion of abdominal effusion (80.5% vs. 54.1%; P < 0.001). Patients in this group demonstrated more severe imaging results, including a higher proportion of bowel wall thickening (97.3% vs. 83.6%, P = 0.001), covering small bowel wall thickening (75.2% vs. 49.2%, P = 0.001), large bowel wall thickening (69.9% vs. 52.5%, P = 0.022), and combined thickening of both small and large bowel walls (48.7% vs. 18.0%, P < 0.001). In addition, the severity of bowel wall thickening was greater in this group (P < 0.001), with a higher proportion of moderate thickening (41.6% vs. 13.1%, P < 0.001) and severe thickening (23.9% vs. 11.5%, P = 0.049). Furthermore, the high-symptom group also had higher proportions of target sign (72.6% vs. 39.3%, P < 0.001) and bowel segment dilation (78.8% vs. 57.4%, P = 0.003) compared with the low-symptom group, while there were no statistically significant differences in the proportions of mesenteric enhancement or increased mesenteric fat density between the two groups. Although no statistically significant differences were observed between the two groups in treatment regimens, more patients in the high-symptom group received very high-dose glucocorticoids (25.7%) (Table 3).

Comparative Analysis of Patients with Symptoms ≤ 7 Days (d) vs. >7 Days (d)

As abdominal pain was the most common symptom for physician visits, 155 patients with acute and chronic abdominal pain duration group tended to choose the long duration group (> 7 d) or the short duration group (≤ 7 d). Compared with long-duration group patients, a lower proportion of short-duration group patients had abdominal distension (48.3% vs. 63.6%, P = 0.058), but they had higher proportions of pericardial effusion (32.6% vs. 15.2%, P = 0.013) and abdominal effusion (78.7% vs. 60.6%, P = 0.014). The comparison of laboratory test results between the two groups showed that WBC count (5.40 vs. 4.15, P = 0.021) and neutrophil count (4.31 vs. 3.07, P = 0.017) of the short duration group were higher than those of the long duration group patients. However, no significant differences were observed in other inflammation-related indicators, including NLR, PLR, SII, ESR, and CRP (Fig. 3). In the two groups, patients in the short-duration group had a higher proportion of bowel wall thickening, particularly combined thickening of both small and large bowel walls (49.4% vs. 31.8%, P = 0.028). Notably, patients in the long-duration group had higher proportions of gastric wall thickening (19.1% vs. 33.3%, P = 0.043) and mild bowel wall thickening (28.1% vs. 48.5%, P = 0.009). No statistically significant differences were observed in other imaging features, such as the target sign, bowel segment dilation, mesenteric enhancement, or increased mesenteric fat density between two groups (Table 4).

Fig. 3.

Laboratory comparison between LMV patients with abdominal pain ≤ 7 days and > 7 days, showing higher WBC and neutrophil counts in the short-duration group

Table 4.

Comparison of Clinical Symptoms and Imaging Features Between the Two Groups

	Abdominal pain > 7d group(n = 66)	Abdominal pain ≤ 7d group(n = 89)	χ2	P值
Clinical symptom[ n (%)]
Fever	27(40.9)	27(30.3)	1.866	0.172
Alopecia	12(18.2)	9(10.1)	2.107	0.147
Skin rash	25(37.9)	22(24.7)	3.106	0.078
Oral ulcer	4(6.1)	4(4.5)		0.724
Arthralgia and myalgia	25(37.9)	25(28.1)	1.662	0.197
Raynaud phenomenon	7(10.6)	10(11.2)	0.015	0.901
Diarrhea	32(48.5)	44(49.4)	0.014	0.907
Abdominal distension	42(63.6)	43(48.3)	3.592	0.058
Nausea and vomiting	53(80.3)	74(83.1)	0.207	0.649
>2 digestive system symptom	48(72.7)	64(71.9)	0.013	0.911
Laboratory test [ n (%)]
Leukopenia	14(21.2)	12(13.5)	1.622	0.203
Thrombocytopenia	10(15.2)	9(10.1)	0.895	0.344
Low complement	41(62.1)	72(80.9)	6.764	0.009
Anti-dsDNA positive	40(60.6)	43(48.3)	2.302	0.129
24-hour urine protein > 0.5 g	24(36.4)	38(42.7)	0.633	0.426
Imaging Results [ n (%)]
Pericardial effusion	10(15.2)	29(32.6)	6.116	0.013
Pleural effusion	36(54.5)	54(60.7)	0.585	0.445
Abdominal effusion	40(60.6)	70(78.7)	5.990	0.014
Multiple serosal effusions	32(48.5)	54(60.7)	2.280	0.131
Bowel wall thickening>5 mm	22(33.3)	17(19.1)	4.077	0.043
Bowel wall thickening>3 mm	63(95.5)	80(89.9)	1.644	0.200
Small bowel wall thickening	42(63.6)	66(74.2)	1.985	0.159
Large bowel wall thickening	40(60.6)	58(65.2)	0.339	0.560
Combined thickening of both small and large bowel walls	21(31.8)	44(49.4)	4.832	0.028
The severity of bowel wall thickening [n(%)]				0.383
Mild(3 ~ 5 mm)	32(48.5)	25(28.1)	6.780	0.009
Moderate(6 ~ 8 mm)	17(25.8)	34(38.2)	2.658	0.103
Sever(>8 mm)	14(21.2)	20(22.5)	0.035	0.851
Target sign	41(62.1)	59(66.3)	0.288	0.592
Bowel segment dilation	47(71.2)	64(71.9)	0.009	0.924
Mesenteric enhancement	66(100.0)	89(100.0)
Increased mesenteric fat density	60(90.9)	79(88.8)	0.188	0.664
Abdominal lymphadenopathy	27(40.9)	49(55.1)	3.035	0.081
Disease activity
SLEDAI	10.0(6.0, 13.0)	8.0(6.0, 11.0)		0.232

SLEDAI: Systemic Lupus Erythematosus Disease Activity Index

Discussion

LMV is a relatively rare but potentially fatal severe complication of SLE involving the digestive system, with an incidence ranging from 0.2% to 9.7% among SLE patients. Clinical manifestations may be protean, ranging from mild to severe. Therefore, many LMV patients are not initially diagnosed in the rheumatology department [3]. In addition, some patients only present with intestinal ischemia in the early stage of the disease without changes in other clinical and immunological indicators, which can easily lead to missed diagnosis and misdiagnosis, resulting in diagnostic delay and adversely affecting patient prognosis [22].

All patients enrolled in our study were newly diagnosed with LMV. Among them, 87 cases (50.0%) had a prior diagnosis of SLE; however, these patients had never presented with gastrointestinal symptoms and thus were not examined with abdominal CT scans. In this study, all patients presented with abdominal pain or other digestive symptoms at diagnosis, suggesting that abdominal imaging was imperative. A previous study has shown that 26.0% of LMV patients developed LMV without a relevant SLE history at the onset of the disease [23]. Given the heterogeneity of LMV and the lack of a comprehensive understanding of the condition, the timely and accurate diagnosis of LMV has become a significant challenge. Based on these observations, we investigated the effects of differences in first-visit departments on clinical symptoms in LMV, with the aim of improving clinicians’ awareness of LMV across departments.

Compared with patients with SLE without LMV, imaging features are the primary basis for diagnosing LMV. In this study, 60.9% of LMV patients presented with the target sign on imaging, which was lower than the rates of 83.5% and 70.0% reported in two previous domestic studies in China. In contrast, no statistically significant differences were observed in other imaging features, including bowel wall thickening, mesenteric enhancement, and increased mesenteric fat density among the three studies [24, 25]. Notably, the incidence of bowel segment dilation varied widely across studies (86.6% vs. 22.0% vs. 71.3% in our study), suggesting substantial heterogeneity in imaging phenotypes. Such inconsistencies may partly reflect differences in admission criteria, sample size, and single-center selection bias; however, geographic and ethnic variations may also contribute. The reported incidence of LMV among SLE patients varies globally (0.2%–9.7%), with Asian populations generally exhibiting higher SLE prevalence and severity than European cohorts [3, 14]. Genetic factors, such as various HLA-DRB1*15 alleles (including SLE risk alleles), may have a role in diverse clinical manifestations in Asian SLE patients, may partially explain the phenotypic heterogeneity observed even within Chinese studies, though its specific link to LMV remains elusive [26, 27]. Direct cross-ethnic comparisons are currently lacking, and future multicenter studies incorporating diverse populations are needed to clarify whether genetic background influences LMV susceptibility and clinical manifestations.

Compared with previous studies, this study had a significantly larger sample size, and its primary objective was to analyze the clinical and imaging features of patients with LMV, all of whom were newly diagnosed. Most previous studies have focused on comparing SLE and LMV patients and identifying risk factors for LMV, but they have not examined differences in the first-visit departments of LMV patients or the effects of the number and duration of digestive symptoms on disease [28]. It should be noted that only 76 cases (43.7%) were initially admitted to the department of rheumatology, whereas 36 cases (20.7%) and 31 cases (17.8%) initially presented to the departments of gastroenterology and emergency medicine, respectively. When comparing symptom profiles between patients who first visited the rheumatology department and those who visited other non-rheumatology departments, patients tended to present to the rheumatology department with relatively specific symptoms, such as arthralgia, myalgia, skin rash, and alopecia. In contrast, the non-specific symptoms of SLE, such as nausea and vomiting, and other digestive system symptoms, were the main reason for approximately 40% of patients’ first visit to other non-rheumatology departments. These patients often had longer hospital stays and a significant increase in their economic burden [7]. In addition, compared with patients who first presented to the rheumatology department, patients who initially visited other departments had lower disease activity and lower levels of inflammation-related indicators (PLR and SII), which may increase medical costs [28, 29]. Among patients for whom treatment plans were formulated, a relatively higher proportion of those who initially presented to non-rheumatology departments did not receive immunosuppressants (17/98). On the one hand, this may be due to differences in disease activity among patients; on the other hand, insufficient awareness of LMV among clinicians in other departments and limited use of immunosuppressants may lead to incomplete consideration of treatment regimens, which may affect patient prognosis [6]. Due to limited awareness of the disease among medical staff in lower-level hospitals and among patients, most patients did not undergo regular follow-up after symptom relief, making it impossible to fully evaluate the impact of treatment regimens from different first-visit departments on prognosis.

Given the heterogeneity of digestive system symptoms in patients with LMV, this study compared the effects of symptom count and duration on clinical laboratory indicators and imaging features, aiming to clarify their impact on patient prognosis. The results showed that the proportions of indicators associated with disease activity, such as leukopenia and hypocomplementemia, were higher in the high-symptom group. In contrast, the proportions of fever and arthralgia with myalgia were higher in the low-symptom group. Based on this, it is speculated that these differences are the primary reason for the absence of statistically significant differences in disease activity between the two groups [11]. When comparing the impact of symptom count on imaging findings, the high-symptom group exhibited more severe imaging manifestations and a higher proportion of abdominal effusion. When analyzing the causes of these differences, it is proposed that digestive system symptoms disrupt internal homeostasis, activate macrophages, and simultaneously induce changes in the gene expression of cytokines and chemokines associated with the NF-κB signaling pathway. In the high-symptom group, the occurrence of multiple symptoms induces the accumulation of numerous macrophages, thereby exacerbating mesenteric artery constriction and reducing intestinal blood flow. Coupled with the stimulation of inflammatory factors, this results in increased intestinal wall permeability, thereby causing abdominal effusion [30]. When comparing the proportions of individual digestive system symptoms, including abdominal pain, abdominal distension, diarrhea, and nausea and vomiting between the two groups, the prevalence of each symptom was higher in the high-symptom group than in the low-symptom group. Previous studies have demonstrated that multiple digestive symptoms, such as abdominal distension, are closely associated with alterations in the intestinal microbiota and metabolic disorders. The release of various oxidative stress products (e.g., reactive oxygen species) and inflammatory factors further stimulates the intestinal wall, promotes persistent intestinal hyper-reactivity, and ultimately leads to bowel wall thickening and dilation [31]. The target sign has been proven to be an important CT sign for predicting small bowel ischemic necrosis. Since intestinal blood flow reduction is more pronounced in the high-symptom group, these patients are more prone to intestinal ischemic necrosis; thus, the target sign is more likely to appear as an imaging feature in this group [32]. In addition, the proportion of patients receiving pulse-dose glucocorticoids was numerically higher in the high-symptom group than in the low-symptom group, and vigilance should be maintained for the occurrence of related adverse reactions, such as gastrointestinal ulcers and bleeding [33]. In summary, when patients present with fewer digestive symptoms, they are more likely to exhibit relatively specific symptoms of autoimmune rheumatic diseases, such as arthralgia and myalgia, fever, fatigue, alopecia, and Raynaud’s phenomenon. Clinicians should maintain high vigilance for these symptoms and signs and promptly screen for SLE. However, when multiple digestive system symptoms occur, despite more severe imaging manifestations, clinicians often fail to detect SLE-specific antibodies because of the absence of SLE-specific clinical symptoms. The above results suggest that when patients present with gastrointestinal symptoms, clinicians should be alert to the possibility of LMV. They should not only perform abdominal CT scans on patients but also comprehensively evaluate the diagnosis of SLE and LMV by integrating laboratory results, including serum complement levels and a full panel of connective tissue disease antibodies. The goal is to administer glucocorticoids and immunosuppressants promptly, implement standardized treatment protocols, and improve patient prognosis.

Abdominal pain is one of the main digestive system symptoms of LMV, and the duration of this symptom is relatively more definite than that of other symptoms. In this study, 155 patients with abdominal pain were divided into the long duration group (> 7d) and the short duration group (≤ 7 d) according to the time period of their abdominal pain, and the impacts of abdominal pain duration on disease activity-related indicators and imaging features were compared between the two groups. Our results showed that patients with long-term abdominal pain, excluding those who visited the rheumatology department, were more likely to seek medical care in the gastroenterology department. In contrast, patients with short-duration abdominal pain preferred emergency medicine. This may be related to the observation that a short-duration group is more similar to acute abdominal pain [14]. This also, to some extent, explains why both the white blood cell count and neutrophil count in patients with abdominal pain lasting ≤ 7 days were higher than those in patients with abdominal pain lasting > 7 days. An increase in neutrophils can promote adaptive immune responses within the abdominal cavity. Studies have shown infiltration of inflammatory CD4 + T cells into intestinal tissues in patients with LMV. Activated inflammatory T cells stimulate innate cells (such as epithelial cells, fibroblasts, and phagocytes), thereby promoting persistent hyper-reactivity to inflammation [34]. Activation of pro-inflammatory adaptive immunity leads to the release of cytokines and chemokines that recruit additional innate immune cells, further impairing intestinal homeostasis. The trafficking of immune cells is a key driver of intestinal inflammation in LMV. Abnormal macrophage polarization occurs during LMV progression, leading to the massive release of pro-inflammatory cytokines. During the inflammatory process, the maturation and differentiation of some monocytes are disrupted, leading to the massive in situ accumulation of TLR-reactive pro-inflammatory macrophages. This induces epithelial cell apoptosis, impairs the intestinal mucosal barrier, and increases intestinal wall permeability, thereby exacerbating abdominal effusion [35, 36]. In addition, the production and accumulation of large amounts of inflammatory factors, coupled with exacerbated mesenteric artery constriction and reduced intestinal blood flow, may induce inferior mesenteric artery occlusion. A recent study has demonstrated that inferior mesenteric artery occlusion is a risk factor for acute mesenteric ischemia, whereas bowel wall thickening is the most common CT manifestation of acute mesenteric ischemia [37]. Therefore, the increased neutrophil counts in patients with abdominal pain lasting ≤ 7 days may partially explain the differences in imaging results. An increase in neutrophils promotes the development of an intra-abdominal inflammatory storm, which reduces intestinal blood flow and alters intestinal wall permeability, thereby inducing abdominal effusion, bowel wall edema, and thickening. In contrast, gastric wall thickening is a relatively slow process; long-term chronic abdominal pain can alter the intestinal flora, particularly with respect to Helicobacter pylori. As a colonizing bacterium in the gut of at least half of the world’s population, Helicobacter pylori infection is closely associated with the development of various types of gastritis and other conditions. In particular, Helicobacter pylori can induce the upregulation of adhesion molecules on gastric epithelial cells and promote the release of cytokines such as IL-1, IL-6, and TNF-α, thereby activating leukocytes and triggering systemic inflammation. Chronic inflammation caused by Helicobacter pylori can precipitate persistent oxidative stress, resulting in a variety of harmful effects (e.g., DNA damage, mitochondrial membrane damage, and pro-inflammatory immune responses), which, in turn, stimulate gastric wall thickening to some extent [38, 39]. Although no statistically significant differences were observed in other imaging features between the two groups, the discrepancies in some laboratory indicators and imaging characteristics caused by variations in abdominal pain duration remind clinicians that when patients present with inconsistent symptoms and signs, they should fully consider multiple possibilities, conduct early screening for SLE and LMV, promptly and accurately identify the underlying nature of the disease, and administer active treatment to avoid severe consequences.

Although, compared with previous studies, this research included a relatively large sample of patients with newly diagnosed LMV and conducted a comprehensive analysis of their clinical and imaging characteristics, it still has certain limitations. First, this study was a single-center, retrospective study. Differences in admission criteria, variations in sample size, and sample selection bias (inherent to single-center research) may be the main causes of inconsistencies in some results [24]. Second, in this study, because clinicians from different departments had different medication experiences, the total doses of glucocorticoids and cyclophosphamide administered to patients during hospitalization were not recorded. This made it impossible to fully evaluate patients’ responsiveness to the drugs. In addition, due to differences in patients’ cognitive abilities and socioeconomic status, some patients did not undergo regular follow-up after symptom resolution. This precluded the assessment of patients’ short-term outcomes and long-term prognosis, including the inability to predict serious disease complications (such as intestinal obstruction, intestinal perforation, and gastrointestinal bleeding), mortality, and disease recurrence. In the future, multicenter, prospective studies with long-term follow-up are needed to further explore prognostic differences among LMV patients and establish reliable disease prediction models.

Conclusion

This study reveals that the non-specificity of clinical symptoms in LMV is closely associated with patients selecting the departments of rheumatology and gastroenterology as their primary and secondary first-visit departments, respectively. However, differences in the choice of first-visit departments may lead to prolonged hospitalization for mild cases, resulting in the waste of medical resources. In addition, there is an association between the number of digestive system symptoms in LMV patients and the severity of their imaging features, suggesting that clinicians should be alert to the occurrence of LMV when patients present with multiple digestive system symptoms. Furthermore, although no significant correlation was observed between abdominal pain duration and disease activity in LMV patients, changes in some laboratory indicators and imaging features induced by varying abdominal pain durations provide important clues for clinicians during the diagnostic process. The results of this study emphasize that in the diagnosis and treatment of patients, clinicians should not be limited to symptomatic treatment; instead, they should comprehensively consider the results of laboratory tests such as serum complement levels and a full panel of connective tissue disease-related antibodies, as well as imaging data such as abdominal CT, to thoroughly explore the underlying causes of the disease. Based on this, a comprehensive diagnosis of SLE and LMV should be made, and standardized treatment with glucocorticoids and immunosuppressants should be promptly initiated to improve prognosis.

Author contributions

All authors contributed to the conception and design of the study. Jiaojiao Han and Shanshan Li contributed equally to this work. The manuscript draft, as well as data collection and analysis, were completed by Jiaojiao Han and Shanshan Li. Pengyu Xie assisted with data collection and analysis. Yujie He and Tianfang Li were responsible for study conception and design. All authors critically reviewed and approved the final manuscript.

Funding

This work was supported by grants from the National Natural Science Foundation of China (grant numbers U1704177 and 81871811).

Data availability

The datasets generated and analyzed during this study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval

This study was approved by the Medical Ethics Committee of the First Affiliated Hospital of Zhengzhou University (Zhengzhou, China) (2025-KY-0428-002).

Consent to publish

All the authors approved the publication.