Clinical characteristics of lymphoma patients presenting with fever of unknown origin misdiagnosed with connective tissue diseases

Abstract

Recognizing and diagnosing lymphoma in patients with fever of unknown origin (FUO) can be challenging, and misdiagnosis is not uncommon. To improve understanding of the clinical characteristics of lymphoma patients presenting with FUO who were misdiagnosed with autoimmune diseases. A retrospective, observational study of 140 consecutive patients with FUO and lymphoma presenting to a tertiary center between January 2017 and December 2023. Patients were divided into those who were correctly diagnosed and those misdiagnosed as connective tissue diseases (CTD) and the clinical features compared. Of 140 lymphoma patients with FUO, 21 patients (15.0%) were misdiagnosed as CTD. The median time between symptom onset and diagnosis was significantly longer in the misdiagnosed group than in the non-misdiagnosed group (11.0 (IQR 6.0, 22.5) months vs. 4.0 (2.5, 9.0) months; p = 0.001). The misdiagnosed group had significantly less lymph node and bone marrow involvement and more skin rashes than the non-misdiagnosed group (47.6% vs. 70.6%, p = 0.039; 23.8% vs. 47.9%, p = 0.040; 47.6% vs. 25.2%, p = 0.036), as well as significantly lower ESR (p = 0.026) and hsCRP (p = 0.049). The misdiagnosed group had higher frequency of ANA/ANCA (57.1% vs. 27.7%; p = 0.008) and anti-phospholipid antibody (42.9% vs. 6.1%; p = 0.008) positivity. The distribution of lymphoma subtypes was different between groups (p = 0.058). Lymphoma patients with an atypical presentation and FUO suggesting inflammatory systemic disease are easily misdiagnosed. Autoantibody positivity is not rare in lymphoma patients with an atypical presentation and FUO, so close follow-up and repeated histopathological examination may be helpful to establishing a correct diagnosis.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00277-025-06188-2.

Introduction

Petersdorf and Beeson defined fever of unknown of origin (FUO) in 1961 as a temperature ≥ 38.3 °C (101 °F) occurring on several occasions, lasting more than three weeks, with a diagnosis that remained uncertain after one week of in-hospital diagnostic workup [1]. The etiological spectrum of FUO is wide and includes infections, non-infectious inflammatory diseases, malignancies, and miscellaneous conditions [2, 3, 4]. Neoplastic causes of FUO, especially lymphoma, are now more common than infectious causes of FUO [5–7]. Diagnosing lymphoma can be difficult due to its complex etiology, lack of characteristic clinical features, and lack of specific biomarkers. Lymphoma patients presenting with FUO are difficult to diagnose with a single biopsy [8–9]. Furthermore, lymphoma is likely to be misdiagnosed as connective tissue disease (CTD) due to the similar presenting symptoms and signs.

Given the difficulties of recognizing and diagnosing lymphoma in patients with FUO, here we retrospectively reviewed the clinicopathological characteristics of lymphoma patients presenting with FUO but either previously correctly diagnosed or incorrectly diagnosed with CTD to explore features that might help with making a correct and timely diagnosis and avoid treatment delay in this severely ill group of patients.

Materials and methods

Patients

This retrospective study was performed at Peking Union Medical College Hospital, a 1800-bed university-affiliated tertiary hospital in Beijing, China. All records of FUO patients diagnosed with lymphoma between January 1, 2017 and December 31, 2023 were reviewed. FUO was defined as an illness with recurrent fever over 38.3 °C lasting three weeks or more and with no established diagnosis after adequate inpatient and outpatient evaluation (a modification of the original one-week diagnostic criterion) [10]. Detailed clinical and demographic information was collected, including the clinical characteristics and physical examination, laboratory, and imaging findings (including ultrasonography, computed tomography (CT), magnetic resonance imaging (MRI) or positron emission tomography/CT (PET-CT) scans) and histopathological evaluations. Clinical features not documented in the medical records were assumed to be absent.

Lymphoma was diagnosed based on histopathological examination and was subtyped according to the 5th edition of the World Health Organization (WHO) Classification of Haematolymphoid Tumors: Lymphoid Neoplasms (2022) [11]. Patients with FUO diagnosed with CTD were excluded from this study. The ethics committee of Peking Union Medical College Hospital approved the study protocol, and all collected data were de-identified.

Statistical analysis

Statistical analyses were performed using IBM SPSS Statistics 20.0 software (IBM Corp., Armonk, NY, USA). Categorical variables are described using counts and percentages. Quantitative variables are described using mean ± standard deviation (SD) or median with interquartile range (IQR). Continuous variables with normal and skewed distributions were compared using Student’s t-test and Wilcoxon rank-sum test, respectively. Categorical variables were compared using χ² tests. A p-value < 0.05 was considered statistically significant.

Results

Patient characteristics

One hundred and forty patients (77 [55.0%] males and 63 [45.0%] females) were included. The average age at diagnosis was 50.2 ± 16.8 years. Patients were divided into two groups based on whether they were misdiagnosed as CTD during the course of the disease at PUMCH, and their demographic and clinical features are presented in Table 1. Twenty-one patients (15.0%) were initially misdiagnosed as CTD. There was no significant difference in the age at diagnosis and sex of the two groups (p = 0.219 and p = 0.794, respectively). For all patients, the median duration prior to diagnosis was 5.0 months (IQR 2.5, 11.0), and 59 (42.1%) patients had experienced symptoms for six or more months. The median time interval between symptom onset and diagnosis was significantly longer in the misdiagnosed group than in the non-misdiagnosed group (11.0 (IQR 6.0, 22.5) vs. 4.0 (IQR 2.5, 9.0) months; p = 0.001), and the misdiagnosed group contained more patients with a disease course exceeding six months (71.4% vs. 37.0%; p = 0.012).

Table 1.

Clinical characteristics of 140 lymphoma patients with FUO

Clinical characteristics	All (n = 140)	Non-misdiagnosed (n = 119)	Misdiagnosed (n = 21)	P-value
Age, years	5 50.2 ± 16.8	51.0 ± 16.5	46.1 ± 18.4	0.219
Sex, male	77 (55.0%)	66 (55.5%)	11 (52.4%)	0.794
Duration, months	5.0 (2.5, 11.0)	4.0 (2.5, 9.0)	11.0 (6.0, 22.5)	0.001
Duration > 6 months	59 (42.1%)	44 (37.0%)	15 (71.4%)	0.012
Time from admission to diagnosis, days	19 (12, 27)	19 (12, 28)	18 (14, 23)	0.700
Fever as the first symptom	88 (62.9%)	80 (67.2%)	8 (38.1%)	0.011
Temperature, °C				0.938
38.4–38.9	41 (29.3%)	35 (29.4%)	6 (28.6%)
≥39.0	99 (70.7%)	84 (70.6%)	15 (71.4%)
With rigors	58 (41.4%)	53 (44.5%)	5 (23.8%)	0.075
Night sweats	23 (16.4%)	20 (16.8%)	3 (14.3%)	0.774
Weight loss > 10%	67 (47.9%)	58 (48.7%)	9 (42.9%)	0.619
Skin rash	40 (28.6%)	30 (25.2%)	10 (47.6%)	0.036
Massively enlarged spleen	100 (71.4%)	86 (72.3%)	14 (66.7%)	0.600
Serous cavity effusion	25 (17.9%)	23 (19.3%)	2 (9.5%)	0.279
Hemophagocytic syndrome	50 (35.7%)	40 (33.6%)	10 (47.6%)	0.217
Extranodal involvement	124 (88.6%)	106 (89.1%)	18 (85.7%)	0.655
More than one extranodal site	69 (49.3%)	59 (49.6%)	10 (47.6%)	0.868
Lymph node involvement	94 (67.1%)	84 (70.6%)	10 (47.6%)	0.039
Liver involvement	21 (15.0%)	19 (16.0%)	2 (9.5%)	0.446
Bone marrow involvement	62 (44.3%)	57 (47.9%)	5 (23.8%)	0.040
Interventions before diagnosis
Antibiotic therapy	112 (80.0%)	98 (82.4%)	14 (66.7%)	0.098
Anti-tuberculosis therapy	21 (15.0%)	17 (14.3%)	4 (19.0%)	0.573
Adrenocortical hormone therapy	72 (51.4%)	51 (42.9%)	21 (100.0%)	< 0.001
Biopsy site selection				0.101
Lymph nodes	41 (29.3%)	38 (31.9%)	3 (14.3%)
Other locations	99 (70.7%)	81 (68.1%)	18 (85.7%)
Biopsy times	2.0 (2.0, 3.0)	2.0 (2.0, 3.0)	2.0 (2.0, 2.5)	0.227
Diagnosis				0.058
Hodgkin lymphoma	17 (12.1%)	17 (14.3%)	0
B-cell NHL	70 (50.0%)	61 (51.3%)	9 (42.9%)
T/NK-cell NHL	53 (37.9%)	41 (34.5%)	12 (57.1%)
Stage III/IV	131 (93.6%)	112 (94.1%)	19 (90.5%)	0.530

Abbreviations: FUO: fever of unknown origin; NHL: non-Hodgkin lymphoma

Overall, fever was the first symptom in 88 patients (62.9%), while other patients experienced skin rash, cough, fatigue, edema, and superficial lymph node enlargement. Fever as a first symptom was significantly less frequent in the misdiagnosed group than in the non-misdiagnosed group (38.1% vs. 67.2%; p = 0.011). Patients in the misdiagnosed group had more skin rash than those in the non-misdiagnosed group (47.6% vs. 25.2%, p = 0.036). There were no significant differences in frequencies of rigors, night sweats, nor weight loss > 10% between the two groups. The frequencies of serous cavity effusion, hemophagocytic syndrome, and extranodal involvement were not significantly different between the two groups. The misdiagnosed group had significantly less lymph node and bone marrow involvement than the non-misdiagnosed group (47.6% vs. 70.6%, p = 0.039 and 23.8% vs. 47.9%, p = 0.040). There was no difference in extranodal and liver involvement between the two groups (Table 1).

Interventions before diagnosis

All patients in the misdiagnosed group had received corticosteroid treatment prior to a definitive diagnosis, which was much higher than the non-misdiagnosed group (100% vs. 42.9%; p < 0.001). There was no difference in antibiotic (66.7% vs. 82.4%; p = 0.098) and antituberculosis (19.0% vs. 14.3%; p = 0.573) prescription between the two groups.

Laboratory and imaging studies

The laboratory findings of the two groups are presented in Table 2. The misdiagnosed group had a significantly lower erythrocyte sedimentation rate (ESR) (37.8 ± 27.9 vs. 54.3 ± 39.3 mm/h; p = 0.026) and high-sensitivity C creative protein (hsCRP) level (33.9 (IQR 12.1, 88.8) vs. 65.1 (IQR 27.4, 116.1) mg/L; p = 0.049). The misdiagnosed group had higher frequency of antineutrophil antibody/antineutrophil cytoplasmic antibody (ANA/ANCA) positivity (57.1% vs. 27.7%; p = 0.008) and antiphospholipid antibody positivity (42.9% vs. 6.1%; p = 0.008) than the non-misdiagnosed group. There were no differences in white blood.

Table 2.

Laboratory and imaging results of 140 lymphoma patients with FUO

Laboratory and imaging	All (n = 140)	Without misdiagnosis (n = 119)	Misdiagnosis (n = 21)	P-value
Pancytopenia, %	19.3% (27/140)	22 (18.5%)	5 (23.8%)	0.569
White blood cell, ×109/L	4.9 (3.1, 7.7)	5.1 (3.1, 7.8)	4.9 (2.8, 8.6)	0.739
Hemoglobin, g/L	98.8 ± 21.5	98.5 ± 21.2	100.7 ± 23.6	0.675
Platelets, ×109/L	127 (74, 233)	127 (74, 233)	128 (75, 246)	0.921
Total bilirubin ≥ 34.2 µmol/L, %	10 (7.1%)	10 (8.4%)	0	0.168
ALT > 40U/L	44 (31.4%)	37 (31.1%)	7 (33.3%)	0.838
Creatinine > 104 µmol/L (male)  >84 µmol/L (female), %	18 (12.9%)	15 (12.6%)	3 (14.3%)	0.832
Albumin, g/L	31.7 ± 5.3	31.7 ± 5.5	31.9 ± 4.5	0.905
ESR, mm/1 h	51.8 ± 38.1	54.3 ± 39.3	37.8 ± 27.9	0.026
hsCRP, mg/L	59.6 (24.7, 115.8)	65.1 (27.4, 116.1)	35.9 (12.1, 88.8)	0.049
LDH, U/L	493 (289, 992)	493 (282, 1002)	460 (331, 798)	0.737
Serum ferritin, ng/mL	912 (423, 2178)	897 (408, 2167)	1310 (505, 2777)	0.300
ANA/ANCA positive, %	45/138 (32.6%)	31/112 (27.7%)	12/21 (57.1%)	0.008
ANA positive, %	39/127 (30.7%)	29/106 (27.4%)	10/21 (47.6%)	0.066
ANCA positive, %	8/114 (7.0%)	6/96 (6.2%)	2/18 (11.1%)	0.459
Anti-phospholipid antibody positive, %	5/40 (12.5%)	2/33 (6.1%)	3/7 (42.9%)	0.008
Anti-ENA antibody positive, %	4/68 (5.9%)	2/55 (3.6%)	2/13 (15.4%)	0.105
EBV-DNA > 400, %	38/135 (28.1%)	31/115 (27.0%)	7/20 (35.0%)	0.460
PET/CT scans				0.729
Positive	85 (60.7%)	73 (61.4%)	12 (57.1%)
Negative	23 (16.4%)	20 (16.8%)	3 (14.3%)
After diagnosis	11 (7.9%)	10 (8.4%)	1 (4.8%)
No results	21 (15.0%)	16 (13.4%)	5 (23.8%)
High uptake of the spleen in PET	79/119 (66.4%)	68/104 (65.4%)	12/16 (75.0%)	0.308

Abbreviations: FUO: fever of unknown origin; ALT: alanine aminotransferase; ESR: erythrocyte sedimentation rate; hsCRP: high-sensitivity C-reactive protein; LDH: lactic dehydrogenase; ANA: antinuclear antibody, ANCA: anti-neutrophil cytoplasmic antibody; ENA: extractable nuclear antigen; PET: positron emission tomography

cell counts and hemoglobin, platelet, lactic dehydrogenase (LDH), and serum ferritin levels between the two groups. There were also no differences in frequency of elevated creatinine, elevated total bilirubin, elevated alanine aminotransferase (ALT), and EBV-DNA positivity between the two groups.

In all patients, 108 patients underwent PET/CT scans before diagnosis, of which 85 patients (78.7%) showed malignant disease. There was no difference in malignant disease on PET/CT between the two groups (73/93, 78.5% vs. 12/15, 80.0%; p = 0.729).

Diagnostic approaches and pathological findings

Biopsy site selection and biopsy time

Patients with FUO in whom lymphoma is suspected should be biopsied to obtain a diagnosis. Several patients required multiple biopsies to ultimately obtain a diagnosis, with a median of two biopsies performed in all patients (range 1–6). There was no difference in the biopsy times between the two groups (p = 0.227).

In the misdiagnosed group, 18 patients (85.7%) underwent extranodal biopsy to obtain a pathological diagnosis, while in the non-misdiagnosed group, the proportion was 68.1%. There was no significant difference between the two groups (p = 0.101).

Histopathological findings

In the misdiagnosed group, twelve patients (57.1%) were diagnosed with T/NK cell non-Hodgkin lymphoma (NHL). The other nine patients (42.9%) were diagnosed with B cell NHL. In the non-misdiagnosed group, 34.5% (41 patients), 51.3% (61 patients), and 14.3% (17 patients) were diagnosed with T/NK cell NHL, B cell NHL, and Hodgkin lymphoma, respectively (p = 0.058).

Clinical characteristics of lymphoma patients in the misdiagnosed group

The clinical characteristics of the 21 misdiagnosed patients are presented in Table 3. Of 21 patients, eight and seven patients were misdiagnosed as connective tissue disease (CTD) and vasculitis, respectively. Other misdiagnoses included dermatomyositis, IgG4-related disease, polymyositis, panniculitis, Sjogren’s syndrome, and eosinophilic fasciitis.

Table 3.

Clinical characteristics of lymphoma patients misdiagnosed as connective tissue disease during the disease course

Patient	Agea/gender	First symptoms	Other symptoms and signs	Biopsy site	Autoantibody positivity	Diagnosis	Misdiagnosis	Treatment prior to diagnosis
1	M/52	Fever	Kidney injury, hematochezia	Ascending colon	ANA 1:80, β2GP1	Extranodal NK/T cell lymphoma	CTD	Glucocorticoids + CTX
2	M/64	Fever, sore throat		Pharyngeal	ANCA	Extranodal NK/T cell lymphoma, nasal type	CTD	Glucocorticoids + CTX
3	F/49	Red rash of right lower limb	Fever, both lower extremities weakness and edema	Kidney	ANCA	Diffuse large B cell lymphoma (non-GCB)	Vasculitis	Glucocorticoids + thalidomide + CTX + CyA
4	F/67	Erythema nodosum of lower limb	Fever, severe concave edema of both lower limbs	Subcutaneous nodule	β2GP1, Acl	Diffuse large B cell lymphoma (ABC)	Vasculitis	Glucocorticoids + CyA
5	M/14	Skin induration and ulceration, fever	Induration in lower limbs, developing into buttocks, upper limbs, and back	Skin	(-)	Extranodal NK/T cell lymphoma	Vasculitis	Glucocorticoids + mycophenolate mofetil
6	M/21	Maxillofacial swelling	Fever, multiple rashes on limbs and trunk	Skin	(-)	Subcutaneous panniculitis-like T cell lymphoma	CTD	Glucocorticoids
7	F/48	Fever		Lymph node	ANA 1:80	Angioimmunoblastic T-cell lymphoma	CTD	Glucocorticoids + CyA
8	F/62	Fever		Lymph node	(-)	Marginal zone B cell lymphoma	Vasculitis	Glucocorticoids
9	M/25	Rash on both calves, edema on both lower limbs	Limb weakness, intermittent fever, cough, pericardial effusion	Bone marrow	(-)	Peripheral T cell lymphoma	Dermatomyositis	Glucocorticoids
10	M/61	Limb weakness	Intermittent fever, pancytopenia	Bone marrow	ANA 1:80, β2GP1	Peripheral T cell lymphoma	CTD	Glucocorticoids + Tacrolimus
11	M/48	Fever		Lung	(-)	Pulmonary intravascular large B-cell lymphoma	CTD	Glucocorticoids + CTX
12	M/52	Rash	Blisters and ruptures on the lower limb, maculae on the face, trunk and upper limb, fever	Skin and muscle	(-)	Peripheral T-cell lymphoma	Vasculitis	Glucocorticoids + thalidomide
13	M/44	Fever	Multiple follicular papules on the lower jaw, chest, abdomen, perineum, and proximal limbs	Bilateral turbinates	AMA-M2	Extranodal NK/T-cell lymphoma	IgG4-RD	Glucocorticoids
14	M/64	Fever	Reduced activity tolerance	Skin and lung	ANA 1:80	Diffuse intravascular large B-cell lymphoma	Polymyositis	Glucocorticoids
15	F/18	Rash on both lower legs	Right foot prolapse, fever	Skin and turbinates	ANA 1:160, anti-C1q	Nasal NK/T-cell lymphoma	Vasculitis	Glucocorticoids
16	F/22	Multiple subcutaneous nodules, fever	Rash on both lower limbs, facial and eyelid swelling, concave edema of both lower limbs	Subcutaneous mass	Anti-Jo1	Subcutaneous panniculitis-like T-cell lymphoma	Panniculitis	Glucocorticoids + CyA
17	F/61	Nasal obstruction, multiple nodules in both lungs	Intermittent fever, dry cough, short of breath	Lung	(-)	Diffuse large B-cell lymphoma, EBV+	Granulomatous vasculitis	Glucocorticoids + CTX
18	F/52	Dry mouth	Vision impairment, fever, chest distress, edema below left knee	Bone marrow	ANA 1:160, ACA 1:160, anti-centromere IgG+, AMA-M2	Diffuse large B-cell lymphoma (non-GCB)	CTD	Glucocorticoids
19	M/21	Sore throat	Fever, fatigue	Epiglottis	ANA 1:80, Anti-Jo1	Extranodal NK/T-cell lymphoma, nasal type	CTD	Glucocorticoids
20	F/48	Fever	Edema in the limbs and face, especially in the lower limbs	Nasopharynx	ANA 1:80, anti-ENA	Diffuse large B-cell lymphoma	Sjogren’s syndrome	Glucocorticoids
21	F/74	Whole-body edema	Multiple bruises on the chest, bilateral groins, popliteal fossa, anterior tibia and ankles, fever	Lymph node	(-)	Diffuse large B-cell lymphoma, EB virus (+)	Eosinophilic fasciitis	Glucocorticoids + CTX + TWP

^aAge at diagnosis

Abbreviations: CTD: connective tissue disease, CTX: cyclophosphamide; IgG4-RD: IgG4 related disease, TWP: Tripterygium wilfordii polyglycoside

Discussion

FUO is a common manifestation of several diseases, but it can still be a perplexing clinical scenario. Patients with FUO are clinically challenging as they are often severely ill with multi-system involvement. However, securing a diagnosis in these patients is not easy: non-invasive measurements such as autoimmune antibodies, serum pathogens, and radiographic imaging are often insufficient to yield a diagnosis. To clarify the clinical characteristics of patients with FUO caused by lymphoma to avoid misdiagnosis, here we retrospectively reviewed and analyzed 140 lymphoma patients presenting with FUO. Twenty-one of our lymphoma patients had been misdiagnosed as CTD during the disease course, accounting for 15.0% of the cohort. These patients had more skin rashes, less lymph node and bone marrow involvement, lower levels of inflammatory indicators, and a higher proportion of autoantibody positivity.

Only 38.1% misdiagnosed patients had fever as the initial symptom, much lower than in the non-misdiagnosed group. Initial symptoms in the misdiagnosed group included rash, skin induration and ulceration, sore throat, and maxillofacial swelling, which happen to be common symptoms of autoimmune disease. Distinguishing between lymphoma and CTD can certainly be challenging due to overlap in clinical symptoms (fever, night sweats, skin rash), leading to misdiagnosis.

CTD are a diverse and clinically heterogeneous set of multisystem diseases caused by abnormal immune responses to multiple genetic and environmental factors [12]. Lymphoma and CTD share pathoetiological and mechanistic factors. Both lymphoma and CTD can involve multiple systems, where they mimic each other, making their diagnosis challenging. Lymphoma often presents with enlarged lymph nodes but, in the present study, lymphoma patients in the misdiagnosed group had fewer lymph nodes involved than the non-misdiagnosed group. The absence of lymph node involvement might be a major reason for misdiagnosis in these patients.

Therefore, in a proportion of cases, lymphoma may mimic CTD, e.g. vasculitis, so patients diagnosed with CTD must be adequately investigated for lymphoma and lymphoma must be ruled out. In this study, lymphoma patients misdiagnosed as CTD had multiple organ involvement and misleading presentations, such as rash and edema (Table 3). For example, case 9 had rashes on both calves and limb weakness, which might be misdiagnosed as dermatomyositis. In other instances, the lymphoma manifested as vasculitis (case 15) or panniculitis (case 16). A variety of autoantibodies have been described in lymphoma [13–14]. But these autoantibodies are not disease-specific (as shown in Table 3). Lymphoma must also be considered when investigating fever with an atypical presentation suggesting inflammatory systemic disease and in the presence of autoantibodies. In the meantime, since CTD may precede the lymphoma, patients diagnosed with CTD also need long-term follow-up.

The clinical presentations of lymphoma are diverse and often atypical, delaying the diagnosis. In the present study, the median time interval between symptom onset and diagnosis was significantly longer in the misdiagnosed group than in the non-misdiagnosed group, and patients in the misdiagnosed group had disease courses exceeding six months. Since patients in the misdiagnosed group were initially misdiagnosed with CTD and received steroids, this improved the lymphoma-related symptoms and masked the diagnosis. Indeed, during corticosteroid tapering, symptoms often recurred or new symptoms appeared, leading to the exclusion of CTD on further investigation. This also suggests that for FUO patients, making a diagnosis does not mark the end of clinical follow-up and that patients should be followed closely and the diagnosis reconsidered in a timely manner as the clinical picture progresses. This study, to our best knowledge, is the first to discuss the misdiagnosis of lymphoma patients presenting with FUO.

PET/CT contributed to the etiological diagnosis in 16–69% of all patients with FUO [16–17]. We found that CT and PET/CT were not specific enough to differentiate CTD from lymphoma, as hypermetabolic lesions can be observed in both conditions. In fact, patients with FUO often need repeat biopsies, as here. Therefore, PET/CT might be useful for decreasing the false negative rate of biopsies and increasing the diagnostic rate of FUO patients with lymphoma by indicating the best biopsy site [8, 15].

This study has several limitations. First, this was a retrospective study conducted in a single center. It is possible that some important epidemiological and clinical details were not recorded or missed. Second, the sample was small and included unbalanced group sizes, which might limit the power of the study and confidence in the results. Finally, this article did not obtain the outcome of the cases and was unable to compare the outcomes of the two groups of patients. Nevertheless, this study still provides useful guidance on how to avoid misdiagnosis of this high-risk group of patients.

Conclusions

This is the first study to describe in detail the clinical characteristics of patients with FUO and lymphoma misdiagnosed as CTD. Lymphoma patients with an atypical presentation and FUO suggesting inflammatory systemic disease are easily misdiagnosed. Autoantibody positivity is not rare in this group of patients, delaying an accurate diagnosis and prompting inappropriate treatment. Therefore, close follow-up and repeated histopathological examination may be helpful to establish a correct diagnosis.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

FUO

Fever of unknown origin

ESR

Erythrocyte sedimentation rate

hsCRP

High-sensitivity C creative protein

ANA

Antineutrophil antibody

ANCA

Antineutrophil cytoplasmic antibody

CT

Tomography

MRI

Magnetic resonance imaging

PET-CT

Positron emission tomography

SD

Standard deviation

IQR

Interquartile range

PUMCH

Peking Union Medical College Hospital

LDH

Lactic dehydrogenase

ALT

Alanine aminotransferase

NHL

Non-Hodgkin lymphoma

CTD

Connective tissue disease

Author contributions

XC and YJ designed the study; XZ, QZ, YW and XH collected data and analyzed the results; XZ and QZ drafted the manuscript; XC and YJ provided supervision or mentorship; all authors revised manuscript critically for important intellectual content. All authors read and approved the final manuscript.

Funding

National High Level Hospital Clinical Research Funding (2022-PUMCH-A-017) and (2022-PUMCH-B-045), and China Medical Board Open Competition Program(20–384)provided the financial support for polishing the language in the article and publication.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The study adhered to the ethical standards laid down in the 1964 Declaration of Helsinki and its subsequent revisions. Ethical approval and waivers of informed consent were obtained from the Peking Union Medical College Hospital Ethics Committee (K2045), and all data collected were de-identified.

Competing interests

The authors declare no competing interests.