Neurosyphilis patients presenting with ocular motor cranial nerve palsy in the modern antibiotic era: Case series from two centers and review of the literature

Abstract

Background

Syphilis can mimic the clinical features of many neurological disorders and cause ocular motor cranial nerve palsy (OMCNP). This study aims to characterize the clinical features of neurosyphilis patients presenting with OMCNP. Methods In a retrospective study and literature review, demographic data, clinical presentations, laboratory findings, radiological findings, treatment approaches, and outcomes were recorded and analyzed.

Results

Overall, 0.74 % and 1.64 % of patients presenting to two institutes exhibited syphilitic OMCNP, respectively. A total of 32 patients (23 from the literature) were recruited, with a median age of 46 years and a predominance of males (81.3 %). Meningovascular neurosyphilis was diagnosed in 29 patients (90.6 %), and tabes dorsalis was found in 3 patients (9.4 %). Unilateral OMCNP was present in 24 patients (65.5 %). Unilateral sixth nerve palsy was observed in 10 patients (34.5 %), followed by unilateral third nerve palsy in 9 patients (31.0 %), and unilateral multiple OMCNP in 5 patients (17.2 %). A total of 12 patients (41.4 %) presented with isolated OMCNP, and 4 of these individuals developed other neurological deficits during the disease. Pupillary involvement was found in all patients who presented unilateral third nerve palsies. After treatment, 17 patients (58.6 %) achieved complete recovery of ocular motility; 6 patients (20.7 %) achieved partial recovery; 5 patients (17.2 %) remained stable; and 1 patient (3.5 %) became progressive.

Conclusions

Patients presenting with OMCNP alone or combined with other cranial nerve deficits should be screened for neurosyphilis, and the standard treatment regimen should be initiated if a diagnosis of neurosyphilis is established.

Introduction

The oculomotor (cranial nerve III), trochlear (cranial nerve IV), and abducens (cranial nerve VI) nerves are the ocular motor cranial nerves, which provide innervation to the extraocular muscles. Ocular motor cranial nerve palsy (OMCNP) is a common disorder in neurological practice and can be a presenting sign of many neurological diseases. The patients may not be able to move their eyes properly and have double vision. They may also experience droopy eyelids and dilated pupils. An etiological study of 4278 cases of OMCNP published in 1992 found that intracranial neoplasm, head trauma, and vascular disease were the three most common causes. In contrast, nearly one-quarter of the other 1037 cases remained “undetermined” (Richards et al., 1992). Syphilis is a systemic infectious disease caused by the spirochete Treponema pallidum, which can spread to the central nervous system within days after exposure and cause neurosyphilis (Ghanem, 2010). Treatment of neurosyphilis is more extensive than that of syphilis without neurological involvements; therefore, recognition of neurological manifestations is clinically important. Syphilis was reported as the cause of 12 of 130 (9.2 %) patients with isolated oculomotor nerve palsy and 10 of 104 (9.6 %) patients with isolated abducens nerve palsy in studies in the 1960s (Green et al., 1964, Shrader and Schlezinger, 1960). After being considered eradicated for many years due to the extensive availability of penicillin, syphilis infection has reemerged in recent years, especially in areas with epidemic levels of HIV infection (Ghanem, 2010, Fisman, 2007, Fenton et al., 2008). This reappearance of syphilis infection in the population reintroduces the importance of recognizing the role of this “old acquaintance” in neurological disorders. In addition, “the great imitator,” syphilis, can mimic the clinical features of many neurological diseases and cause OMCNP or other eye movement disorders, further complicating diagnosis. Early neurosyphilis can be asymptomatic or present with meningitis and meningo-vasculitis. Late neurosyphilis includes general paresis and tabes dorsalis. The OMCNP may be due to small vessel vasculitis and nerve infarction in syphilitic meningitis or gumma, occurring during any stage or be associated with neurosyphilis (Ghanem, 2010). The host immunity is important in the course of syphilis. Several studies have demonstrated the role of B cells and CXC chemokine ligand −13 (CXCL13) in the pathophysiology of neurosyphilis (Marra, 2021). The different strains of Treponema pallidum and host genetics may also play a role in developing neurosyphilis (Sahi et al., 2021). So far, most cases in the modern antibiotic era are reported sporadically. Based on literature review and our data, the study aimed to improve our understanding of the clinical features, diagnosis, and treatment of OMCNP caused by neurosyphilis in the modern antibiotic era.

Methods

Patients

In the present study, the medical charts of all patients admitted to the Department of Neurology of Beijing Tongren Hospital between January 2009 and December 2012 and the Department of Ophthalmology of Renji Hospital, Shanghai Jiao Tong University, School of Medicine between June 2015 and December 2022, with a diagnosis of any OMCNP and “neurosyphilis” were reviewed. In addition, the demographic features, past medical history, presenting and accompanying symptoms, physical examinations, blood and/or cerebral spinal fluid (CSF) laboratory tests, image results, and follow-up data were documented. The inclusion criteria for this study were that patients: 1. Presented mainly with ophthalmoplegia; 2. Met the diagnostic criteria of “confirmed neurosyphilis” established by the Centers for Disease Control (CDC); (Workowski et al., 2021a) 3. Had no history, laboratory, or image evidence of any other infectious, metabolic, connective tissue disorders or neoplastic disorder; and 4. Had at least one follow-up visit after treatment. This study was performed in line with the principles of the Declaration of Helsinki. Informed patient consent was obtained from all included patients. The Capital University of Medical Sciences, Beijing Tongren Hospital and Renji Hospital, Shanghai Jiao Tong University, School of Medicine Review Board approved this retrospective study. Relevant clinical details were extracted from patients’ medical charts.

Case presentation

Here we describe one representative patient (Case 25 in Table 1).

Table 1.

Clinical features, laboratory tests, image study, treatment, and outcome of 32 cases of syphilitic ocular motor cranial nerve palsies.

Case No.	Author, year	Age(y)	Gender	Disease duration	Presenting affected nerve/laterality	Pupil involved*	Other lesions developed	Type	Blood serology	CSF serology	MRI/CT	Biopsy	Treatment	Outcome
1	Jordan,1978	61	M	2d	III/Bi	Y	None	MV	VDRL 1:8	VDRL 1:1	NA	NA	PP. I.M., BP. I.M.†	Resolved
2	Kplan,1981	28	M	2wks	III/L	Y	Retro-orbital pain	MV	VDRL(+)	FTA(+)	Gumma Frontal lobe	Plasma cells Granuloma	P. I.M. 3wks‡ 2 course	Partial recovery
3	Currie 1988	58	M	37 yrs§	III/Bi, VI/L	Y	Hearing loss Ataxia ON Headache	MV	VDRL(-)	(-)	Gumma R.PCA Atrophy	Caseation necrosis Plasma cells	P 24MU/d I.V. 3wks	Stable
4	Currie 1988	41	M	7 yrs||	III/L, VI/L	Y	Hemiparesis Horner Headache Hearing loss	MV	VDRL(-) FTA(+)	(-)	Lacunar infarction Atrophy Gumma L.PCA	Lymphocytes Endarteritis Perivasculitis	PP. 0.6MU/d I.M. 15d, repeat therapy	Resolved
5	Slavin,1992	34	M	2wks	VI/R	N	Hearing loss Ataxia	MV	VDRL(-)	VDRL(+)	Gumma Pontine	NA	P. I.V.¶	Resolved
6	Vogl,1993	44	F	7wks	III/R	Y	Frontal headache	MV	VDRL(+)	NA	Gumma Midbrain	NA	P. 30MU/d I.V. 2wks	Resolved
7	Inoue, 1995	51	M	3 m	VI/R	N	Headache Ataxia Vertigo	MV	TPHA(+)	TPHA(+)	Gumma T. lobe Cerebella infarction	Caseation necrosis Plasma cells	P. 12MU/d I.V. 10d	Stable
8	Stepper,1998	38	M	2 yrs	III/Bi, VI/L	Y	Ataxia	TD	VDRL 1:64	VDRL 1:4	Dorsal column increased T2	NA	P. I.V. 3 wks¶	Stable
9	Nadgir,2003	50	M	2wks	CS/R	Y	Headache	MV	VDRL 1:64	VDRL(+)	Mass in CS	NA	P 24MU I.V. 2wks	Resolved
10	Corr, 2004	22	F	4wks	III/Bi	Y	Headache	MV	NA*,*	FTA(+) VDRL(+)	Enhanced, enlarged III/Bi	NA	P. I.V.¶	Resolved
11	Seeley, 2004	54	M	4d	III/R	Y	None	MV	RPR 1:64	VDRL 1:8	Enhanced, enlarged III/R	NA	P. 24MU/d I.V. 2wks	Resolved
12	Jacob,2005	56	M	2.5 yrs	III/Bi, VI/Bi, IV/R	Y	ON/Bi	MV	VDRL 1:64	VDRL 1:2	Brachiocephalic trunk aneurysm††	NA	P. 24MU/d I.V. 2wks	Stable
13	Noel, 2011	62	F	2wks	CS/L	Y	Pain Aortic murmur	MV	VDRL 1:16	VDRL 1:16	Enhanced mass in CS	NA	P. I.V. and P. I.M.‡‡	Partial recovery
14	Rolimpandoei, 2018	57	M	25d	III/L	Y	None	MV	RPR 1:4 TPPA 1:640	RPR 1:1	Multiple patchy lesions	NA	P. I.V. and P. I.M.‡‡	Partial recovery
15	Rolimpandoei, 2018	45	F	1d	III/L, VI/L	Y	Itchy Maculopapular rash	MV	RPR 1:32 TPPA 1:640	RPR 1:2	Normal	NA	P. I.V. and P. I.M.‡‡	Resolved
16	Kowalski, 2019	65	M	1d	VI/L	N	Reduced visual acuity Optic nerve atrophy	MV	RPR 1:128	TPPA(+)	Normal	NA	BP 10.8 g/d I.V. 2wks	Progressive
17	Singh, 2020	34	M	3d	VI/L	N	Headache Mild altered mental status	MV	RPR 1:8	VDRL 1:16	Normal	NA	P. 12MU/d I.V. 2wks	Partial recovery
18	Nes, 2021	30 s	M	2d	III/L	N	Reduced visual acuity	MV	TPPA 1:480 RPR 1:256	RPR(+)	Enlarged lymph nodes in mediastinum and above the neck	NA	BP I.V.¶	Resolved
19	Antaki, 2021	47	M	2wks	III/R	Y	Headache	MV	RPR 1:16	VDRL 1:2	Enhanced III/R	NA	P. 24MU/d I.V. 2wks	Partial recovery
20	Zhou, 2022	36	M	NA	NA	N	NA	MV	TRUST 1:32	TRUST 1:4	Normal	NA	NA	NA
21	Zhou, 2022	35	M	NA	NA	N	NA	MV	TRUST 1:128	TRUST 1:8	Normal	NA	NA	NA
22	Zhou, 2022	32	M	NA	NA	N	NA	MV	TRUST 1:4	TPHA(+)	Normal	NA	NA	NA
23	Garcia, 2022	33	M	1 m	VI/L	N	Headache	MV	RPR 1:128	RPR(+)	Normal	NA	P. 24MU/d I.V. 2wks	Partial recovery
24	Zhu, 2023	44	M	30d	III/R	Y	None	MV	RPR 1:16	RPR 1:4	Normal	NA	Ceft.2 g/d I.V. 2wks	Resolved
25	Zhu, 2023	44	M	2 m	III/L	Y	Ataxia	TD	RPR 1:8	RPR 1:2	Enhanced III/L	NA	Ceft.2 g/d I.V. 2wks	Resolved
26	Zhu, 2023	56	M	40d	III/R, VI/R	Y	ON/Bi	MV	RPR 1:128	RPR 1:8	Normal	NA	P. I.V. and P. I.M.‡‡	Stable
27	Zhu, 2023	53	F	40d	VI/L	N	None	MV	RPR 1:128	RPR 1:8	Normal	NA	P. I.V. and P. I.M.‡‡	Resolved
28	Zhu, 2023	57	F	2wks	VI/R	AR	Lower limb weakness	TD	TRUST 1:32	VDRL 1:8 TRUST 1:4	Normal	NA	P. 24MU/d I.V. 2wks	Resolved
29	Zhu, 2023	64	M	4 m	VI/R	N	None	MV	TRUST 1:32	VDRL(-) TRUST(-)	Normal	NA	P. 24MU/d I.V. 2wks	Resolved
30	Zhu, 2023	62	M	2 m	VI/L	N	None	MV	TRUST 1:128	VDRL(+)	Normal	NA	P. 24MU/d I.V. 2wks	Resolved
31	Zhu, 2023	52	M	2 m	III/L	AR	Reduced visual acuity	MV	TRUST 1:64	VDRL 1:2 TRUST 1:1	Normal	NA	P. 24MU/d I.V. 2wks	Resolved
32	Zhu, 2023	42	M	1wk	VI/R	N	None	MV	TRUST 1:32	TRUST 1:2 VDRL 1:4	Normal	NA	Ceft.2 g/d I.V. 2wks	Resolved

Abbreviations: AR, Argyll-Robertson pupil; Bi, bilateral; BP, benzathine penicillin; Ceft, ceftriaxone; CS, cavernous sinus; d, days; F, female; FTA, fluorescent treponemal antibody absorption test; I.M., intramuscularly; I.V., intravenously; L, left; M, male; m, months; MU, million units; MV, meningovascular syphilis; NA, not available; ON, optic neuropathy; P, penicillin; PCA, pontine-cerebellar angle; PP, procaine penicillin; R, right; RPR, rapid plasma reaction; T, temporal; TD, tabes dorsalis; TPHA, treponema pallidum hemagglutination assay; VDRL, venereal disease research laboratory; wks, weeks; yrs, year

^*The pupil involvement only included patients with enlarged pupil caused by cranial nerve III palsy, Argyll-Robertson pupil had been additionally noted, ^†Intramuscular procaine penicillin 1.2MU per day for 2 weeks and intramuscular benzathine penicillin 2.4MU per week for 8 weeks, ^‡Intramuscular penicillin 7.2MU per week, ^§Duration of recurrent ophthalmoplegia before the syphilitic gumma was confirmed by biopsy, ^||Case 4 experienced eight episodes of ophthalmoplegia in a period of 16 years, and only the first course, when the diagnosis of gumma was reached by biopsy, is included in our review analysis, ^¶The doses of penicillin were not available in the article, ^**Serum antibody test for human immunodeficiency virus was positive, ^††Normal orbital CT and MRI and MRA, ^‡‡Intravenous aqueous penicillin G 24 MU per day for two weeks and followed by intramuscular benzathine penicillin 2.4 MU once per week for 3 weeks

In September of 2009, a 44-year-old man presented with a 2-month history of progressive diplopia associated with left ptosis and a 2-year history of progressive difficulty in walking. He had not previously received any diagnosis or treatment. Past history: He reported no history of systemic, neurological, or ocular disease. Examination & Evaluation: His left eye showed mild ptosis and marked limitations of adduction and elevation. His left pupil was round but enlarged to 4 mm and sluggishly reactive to light. He exhibited hypoesthesia, decreased vibration, joint position sensations, and decreased tendon reflexes on both lower extremities. A heel-shin coordination test revealed unsteady coordination, and Romberg’s sign was positive. No other remarkable neurological signs were found. A neostigmine test and repetitive nerve stimulation were both negative. His serum rapid plasma reagin (RPR) (1:8) and Treponema pallidum hemagglutination (TPHA) were positive; however, the results of other extensive laboratory screening, including serum folic acid levels and autoimmune disorder antibody sets, were normal. A CSF examination revealed a normal pressure of 175 mmH₂O, an elevated WBC count of 60×10⁶/L, no red blood cells, a mildly elevated protein level of 56 mg/dl, and a normal glucose level of 3.16 mmol/L. His CSF RPR was positive (1:2) with a positive TPHA. An MRI scan revealed an enhancement of the left oculomotor nerve at the prepontine cistern with otherwise normal cerebral parenchyma (Fig. 1). The thoracic and lumbar spine MRI scans without contrast found no abnormality. A head MRA failed to indicate any evidence of an intracranial aneurysm. Treatment & Prognosis: He was treated with intravenous ceftriaxone sodium (2 g/d) for two weeks, and the impaired movement of the left eye significantly improved. At the six-month follow-up evaluation, clinical manifestations were stable, but CSF analysis revealed a reactive RPR, and intravenous treatment with ceftriaxone was administered for an additional two weeks. At another follow-up 39 months after the first anti-syphilis treatment, he showed a complete resolution of ocular motility with mild ptosis and a sluggish response of the left pupil. No aberrant regeneration was noted. Unfortunately, a CSF evaluation and blood RPR were not performed at this last follow-up evaluation.

Fig. 1.

Axial (A) and coronal (C) T1-weighted brain MRI scans showing oculomotor nerve at prepontine cistern (arrows). Gadolinium enhanced axial (B) and coronal (D) T1-weighted brain MRI images with fat saturation demonstrating enhanced left oculomotor nerve at prepontine cistern with an otherwise normal cerebral parenchyma (arrows).

Literature review

The literature selection for this review was based on a PubMed database search of publications in English from 1965 to January 2023. The search was performed using various combinations involving the following terms: “(ocular motor nerve palsy OR diplopia OR ophthalmoplegia) AND (neurosyphilis OR syphilis).” Relevant papers were then cross-referenced from the articles cited by the identified studies. Ultimately, 23 patients with neurosyphilis who also presented with OMCNP were identified from the literature review of PubMed citations and are shown in Table 1 (Kaplan et al., 1981; Jordan et al., 1978; Slavin et al., 1992; Vogl et al., 1993; Inoue et al., 1995; Stepper et al., 1998; Corr et al., 2004; Seeley and Venna, 2004; Jacob et al., 2005; Currie et al., 1988; Nadgir et al., 2003; Noel et al., 2011; Nes and Lif Breivik, 2021; Garcia et al., 2022; Zhou et al., 2022; Rolimpandoei et al., 2018; Kowalski et al., 2019; Antaki et al., 2021; Singh et al., 2020).

Results

Frequency of syphilitic OMCNP in our institute

From January 2009 to December 2012, 537 patients who presented with OMCNP were examined at the Department of Neurology of Beijing Tongren Hospital. Overall, four (0.74 %) patients exhibited syphilitic OMCNP. From June 2015 to December 2022, 305 patients who presented with OMCNP were examined at the Department of Ophthalmology of Shanghai Renji Hospital. Overall, five (1.64 %) patients exhibited syphilitic OMCNP. A total of 9 new patients with syphilitic OMCNP were identified during our chart review; 3 patients whose records did not include brain magnetic resonance imaging (MRI) capable of entirely ruling out other possible causes of OMCNP were excluded.

Characteristics of syphilitic OMCNP

A total of 32 cases (23 cases from the literature review and 9 cases from our data) were analyzed, and the patient characteristics are described in Table 2 (case 20–22 only had information about patients’ age, gender, type of neurosyphilis, blood and CSF results and MRI findings, so the calculation afterwards didn’t include the missing data in these 3 cases). Demographic data: The study included 26 males (81.3 %) and 6 females (18.7 %). The median age was 46 years, ranging from 22 to 65. Topographic findings: Unilateral OMCNP was present in 19 patients (65.5 %), with unilateral cranial nerve III palsy in 9 patients (31.0 %), unilateral cranial nerve VI palsy in 10 patients (34.5 %). Unilateral multiple cranial nerves palsies in 5 patients (17.2 %). Two patients (6.9 %) presented with bilateral cranial nerve III palsies, 2 patients (6.9 %) presented with bilateral cranial nerve III palsies accompanied by cranial nerve VI palsy, and 1 patient (3.4 %) presented with bilateral cranial nerve III palsies accompanied by cranial nerve IV and cranial nerve VI palsies. In 19 patients with cranial nerve III palsy, 18 patients had an enlarged pupil and a decreased light response. One patient with cranial nerve III palsy and 1 patient with cranial nerve VI palsy had Argyll-Robertson pupils. Associating signs and symptoms: Isolated OMCNP was observed in 12 patients (41.4 %), of whom 4 developed other neurological deficits during the disease (ranged from a few months to over 30 years), including optic neuropathy, hearing loss, ataxia, Horner’s syndrome, hemiplegia, myoclonus, memory loss, and asymptomatic cephalic trunk aneurysm. The other 17 patients (58.6 %) exhibited other associated neurological deficits when presenting with OMCNP, including optic neuropathy, hearing loss, ataxia, vertigo, muscle weakness, and cavernous sinus syndrome. In 10 patients (34.5 %), accompanying retro-orbital pain or headache was noted. Laboratory and imaging: Blood and CSF results were available in 31 patients. Syphilis serological tests and cerebrospinal tests were positive in 29 patients (93.5 %) and 28 patients (90.3 %), respectively. All patients showed CSF pleocytosis and elevated protein levels. Two cases with HIV co-infection were reported in the literature, (Corr et al., 2004, Kowalski et al., 2019) and one of our patients presented co-infection with hepatitis C. Brain and/or spine computed tomography (CT) and/or MRI examinations were available in 31 patients, of whom 13 (41.9 %) exhibited abnormalities responsible for OMCNP (Table 1). Treatment and prognosis: All patients received intravenous or intramuscular penicillin or intravenous ceftriaxone treatment. Data of follow-up visits was available in 29 patients, among which 17 patients (58.6 %) achieved complete recovery of ocular motility; 6 patients (20.7 %) achieved partial recovery; 5 patients (17.2 %) remained stable; and 1 patient (3.5 %) became progressive. After anti-syphilis treatment, four patients (13.8 %) experienced single or recurrent episodes of OMCNP.

Table 2.

Summary of clinical features of 32 combined syphilitic OMCNP cases.

Characteristic	Value
Median age at onset, years (range)	46, (22−65)
Sex, No. (%) of all patients
Male	26 (81.3 %)
Female	6 (18.7 %)
Types of neurosyphilis, No. (%) of all patients
Meningovascular neurosyphilis	29(90.6 %)
Tabes dorsalis	3(9.4 %)
Laterality, No. (%) of all patients
Unilateral	24 (82.8 %)
bilateral	5 (17.2 %)
Median duration of ophthalmoplegia when diagnosed, (range)	28 days, (2 days−37 years)
Accompanied retro-orbital pain or headache, No. (%) of all patients	10 (34.5 %)
Recurrent episodes of ophthalmoplegia, No. (%) of all patients	4(13.8 %)
Associated neurological deficits during the whole disease durations, No. (%) of all patients Isolated OMCNP without any other associated neurological deficits	8 (27.6 %)
With associated neurological deficits developed after presentation	4 (13.8 %)
With associated neurological deficits before or at presentation	17 (58.6 %)
Ocular motor nerve(s) involved, No. (%) of all patients
Unilateral III	9 (31.0 %)
Unilateral VI Unilateral multiple	10 (34.5 %) 5 (17.2 %)
Bilateral III	2 (6.9 %)
Bilateral multiple	3 (10.4 %)
No. (%) of pupil involvement in 9 patients with unilateral III nerve palsy	9 (100 %)
Prognosis of ocular motility, No. (%) of all patients
Completely resolved	17 (58.6 %)
Partial resolved	6 (20.7 %)
Stable Progressive	5 (17.2 %) 1(3.5 %)

Abbreviations: OMCNP, ocular motor cranial nerve palsy

Discussion

Diagnosis of neurosyphilis remains challenging due to its multiform and polymorphous clinical and laboratory manifestations and its ability to imitate other diseases. The clinical manifestations of neurosyphilis can be divided into early and late stages. Early neurosyphilis (so-called meningovascular neurosyphilis) can be classified as asymptomatic, symptomatic meningitis, and meningo-vasculitis, while late neurosyphilis (so-called parenchymatous neurosyphilis) includes dementia paralytica and tabes dorsalis. Those who cannot fall into either of the two stages are classified as atypical neurosyphilis. Meningovascular neurosyphilis, which includes both the meningeal and vascular neurosyphilis subtypes, represents approximately 10 % of all cases of neurosyphilis and tends to appear 5–12 years after the initial infection with Treponema pallidum (Ghanem, 2010, Inoue et al., 1995).

According to a review of neurosyphilis, cranial nerve palsy can appear during any stage or be associated with neurosyphilis, from early meningeal syphilis to general paresis and tabes dorsalis (Ghanem, 2010, Inoue et al., 1995). Still, this condition is most associated with the meningovascular type and is rarely linked to parenchymatous neurosyphilis. According to a retrospective review in Alberta, (Landry et al., 2019) Patients with early neurosyphilis are more likely to have ocular involvement than those of late stage. The involvement of the third, fourth, and sixth nerve in neurosyphilis patients can all lead to diplopia. Apart from this, other ocular manifestations, including changes in the pupil, visual acuity, and visual field, could also occur in neurosyphilis patients, which should be paid attention to. We found a predominance of males and an age distribution that mostly ranged from 40 to 60 years in this group of patients. 90.6 % of these patients met the clinical diagnosis criteria for meningovascular syphilis. Only 3 patients received a clinical diagnosis of tabes dorsalis. This distribution was consistent with previous reports and reviews of neurosyphilis cases from the pre-antibiotic era (Ghanem, 2010, Nakashima et al., 1996).

Unilateral oculomotor nerve palsy was found to be the most common clinical type among our neurosyphilis patients who presented with OMCNP. The potential mechanism of isolated oculomotor nerve palsy may be due to small vessel vasculitis and nerve infarction in the meningovascular stage. It may also be due to vigorous granulomatous response of the meninges caused by nerve gumma. Interestingly, these cases mostly presented enlarged and poorly responsive pupils instead of Argyll-Robertson pupils. Usually, pupil involvement during isolated third nerve palsy is a sensitive predictor of aneurysmal compression (Yanovitch and Buckley, 2007). This finding may be due to the compression of the nerve fibers innervating pupils due to meningitis. It may also be because the presentation of Argyll-Robertson pupils more often occurs in the late stage of neurosyphilis. And our patients who presented mainly with ophthalmoplegia were probably in the early stage. Our results further suggest that neurosyphilis should be considered during the differential diagnosis of unilateral oculomotor nerve palsy with pupil involvement after ruling out aneurysm and other common causes of OMCNP. Furthermore, unilateral abducens nerve palsy was also a common presenting symptom in our neurosyphilis patients. The assumed mechanism may be similar to that of isolated oculomotor nerve palsy. Due to the clinical and pathogenic similarities between this condition and microvasculopathic sixth nerve palsy commonly associated with diabetes and hypertension, a unilateral isolated sixth nerve palsy caused by neurosyphilis should be routinely considered in an adult patient without microvascular risk factors, especially when the symptoms last for more than three months. Regarding bilateral oculomotor nerve palsy or cavernous sinus syndrome associated with neurosyphilis, although these conditions are quite rare in our study and previous reports, they must still be considered during differential diagnosis because of their curability.

Approximately 40 % of our neurosyphilis patients who presented with OMCNP were isolated, among whom a few later developed other neurological lesions that led to optic neuropathy, hearing loss, and cerebellar-pontine ataxia. However, only a small subset of patients developed remote lesions causing hemiplegia, memory loss, and other mental disorders during disease durations that lasted for 37 years. Notably, these lesions were mainly located in the front, middle or posterior fossa, with 65.5 % being unilateral. The possible underlying mechanism of this finding is the gradual extension of syphilitic meningovascular inflammation within adjacent structures along the skull base. From a clinical viewpoint, this result should reinforce the need to pay special attention to inflammatory lesions of the structures of the skull base area when examining a patient with OMCNP suspected of being caused by neurosyphilis.

MRI is an important diagnostic tool for lesion localization, especially for small lesions that are difficult to locate using brain CT, including those at the skull base. MRI with contrast can reveal even smaller lesions, provided that they are active. Within our group of 32 OMCNP cases, 31 included data from CT and/or MRI examinations. Interestingly, gummas, which are thought to be a rare manifestation of tertiary neurosyphilis, were quite common (6 of 31, 19.4 %) in our combined group of syphilitic OMCNP patients. Some gummas were causative, and others were asymptomatic but still indirectly suggestive of syphilis. Fargen et al. found that most cerebral gummas were located within gyri and the pituitary area (Fargen et al., 2009). The mechanism of gumma formation is believed to be a localized form of syphilitic meningitis. The inflammatory components during the early stage gradually become fibrotic and ultimately develop into a granuloma. The similar focalized lesions found in patients with neurosyphilitic OMCNP and cerebral gumma further suggest subacute or chronic meningovascular inflammation as the primary pathogenic process of OMCNP caused by neurosyphilis. Four patients showed an enhanced and enlarged oculomotor nerve on brain MRI with contrast, providing direct diagnostic evidence of inflammation in the oculomotor nerve. No other accompanied lesions suggestive of neurosyphilis were found in these four patients. However, three patients diagnosed with neurosyphilis with diplopia as main complaint had normal MRI results in a study, but they did not receive MRI with contrast (Zhou et al., 2022). This study indicated that patients with neurological involvement and normal MRI results should have an mNGS (metagenomic next-generation sequencing) test to rule out neurosyphilis. Together with our study, we suggest that for patients with isolated OMCNP suspected to be a result of neurosyphilis, brain MRI with contrast and mNGS test might be an additional diagnostic tool after a negative routine brain MRI.

Although we adopted a stricter set of diagnostic criteria established by the CDC for our newly diagnosed neurosyphilis patients, two patients from the literature review did not show serum or CSF-positive syphilitic serology (Corr et al., 2004, Currie et al., 1988). These two individuals were diagnosed by ruling out other causes and performing biopsies that revealed central caseating necrosis with plasma cell infiltration; they were then classified as seronegative syphilis. This finding is understandable given the varied sensitivity of the venereal disease research laboratory (VDRL) test and RPR during various stages of syphilis. When diagnosing late neurosyphilis, the VDRL and RPR may be negative in up to 30 % of patients (Ghanem, 2010). The CSF treponemal assay, like CSF TPPA and CSF fluorescent treponemal antibody (FTA) test, are suggested to reduce the false-negative rate for those highly suspected of neurosyphilis (Ghanem, 2010). Nowadays, new automated treponemal tests such as reverse syphilis algorithm and transcription-mediated amplification assay were also recommended for their good sensitivity and specificity (Getman et al., 2021, Rourk et al., 2016).

The recommended regimen for neurosyphilis is intravenous aqueous crystalline penicillin G (18–24 MU) for 10–14 days, followed by intramuscular benzathine penicillin (2.4 MU) once per week for up to 3 weeks (Workowski et al., 2021a). Limited data suggested that an alternative ceftriaxone treatment could yield acceptable results in patients allergic to penicillin (Workowski et al., 2021b). Some studies even reported that the patients with cerebral gumma improved with steroid treatment alone, and a spontaneous improvement of diplopia and ophthalmoplegia was observed in some patients (Stepper et al., 1998, Currie et al., 1988, Fleet et al., 1986, Herrold, 1994). Not all of our combined patients finished the recommended treatment regimen following CDC guidelines due to allergies to penicillin or poor compliance. After treatment, 23 of 29 (79.3 %) patients achieved complete or partial recovery of their ocular motor symptoms, and 7 of 29 (24.1 %) patients remained stable. It indicated that if treated timely and adequately, the ocular motor symptoms of neurosyphilis can recover. Also, some of our patients experienced a recurrence of impaired ocular motility, so regular follow-up and timely detection of the recurrence symptoms are essential for neurosyphilis patients. Besides, serofast is a significant concern in treating neurosyphilis, which means a lack of serological response to anti-syphilitic therapy (Clement et al., 2014). The mechanism of serofast is still unclear. Some studies believed comorbid HIV or other immune diseases might cause it, and the condition of case 16 supported this idea. It is worth pointing out that the neurosyphilis patients increased with the rising incidence of HIV infection (Ghanem, 2010). Either syphilis or HIV can make the individual more susceptible to the other, and the comorbidity of the diseases can lead to faster progression and worse prognosis, which raise our attention to detect the underdiagnosed disease as early as possible.

To our knowledge, our group of 9 new neurosyphilis cases who also presented with OMCNP is the largest case series since 1965, and it only includes approximately 1 % of all OMCNP patients. However, the rarity of this disease does not decrease the importance of diagnosis because permanent damage can occur without early treatment. A limitation of our study was that not all our 9 new patients underwent a brain MRI with contrast, which might cause some inaccurate descriptions of image changes due to the omission of small active lesions, but this omission should not affect the final diagnosis of neurosyphilis.

In summary, isolated oculomotor nerve and/or abducens nerve palsies were the most common presenting signs of OMCNP caused by neurosyphilis, and a subset of patients developed additional neurological lesions around the skull base structure. Combining serological tests for syphilis and brain MRI scans with and without contrast can provide good diagnostic evidence and evaluate treatment outcomes. As the epidemic of syphilis reemerges, any patient presenting with OMCNP alone or in combination with other cranial nerve deficits should be screened for neurosyphilis, particularly in the absence of conventional microvascular risk factors, such as diabetes mellitus. The standard treatment regimen should be initiated once a diagnosis of neurosyphilis has been established.

Ethics statement

The study was conducted in accordance with the principles of the Declaration of Helsinki, and approved by the ethics committee of Beijing Tongren Hospital and Shanghai Renji Hospital. The procedures used in this study adhere to the tenets of the Declaration of Helsinki.

Consent

Informed consent was obtained from all individual participants included in the study. Additional informed consent was obtained from all individual participants for whom identifying information is included in this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Author statement

YY and XZ contributed the conception of the study and the clinical image analysis. SC, JP and HJ contributed to MR data acquisition. MZ, SC, JP, HJ, YY and XZ contributed the clinical data acquisition. MZ, SC, JP, HJ, YY and XZ contributed to the interpretation and draft. MZ, SC, JP, HJ, YY and XZ contributed to the revision for important intellectual content.

CRediT authorship contribution statement

Mingjie Zhu: Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing. Shilei Cui: Data curation, Formal analysis, Investigation, Methodology, Resources, Writing – original draft, Writing – review & editing. Hanqiu Jiang: Data curation, Formal analysis, Investigation, Methodology, Resources, Writing – original draft, Writing – review & editing. Yan Yan: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing. Jingting Peng: Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing. Xiaojun Zhang: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing.

Declaration of Competing Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.