Article Text

Clinical spectrum of late symptomatic neurosyphilis in China: an 11-year retrospective study
  1. Yilan Yang,
  2. Xin Gu,
  3. Lin Zhu,
  4. Yuanyuan Cheng,
  5. Haikong Lu,
  6. Zhifang Guan,
  7. Mei Shi,
  8. Liyan Ni,
  9. Rui-Rui Peng,
  10. Wei Zhao,
  11. Juan Wu,
  12. Tengfei Qi,
  13. Fuquan Long,
  14. Zhe Chai,
  15. Weiming Gong,
  16. Meiping Ye,
  17. Pingyu Zhou
  1. Institute of Sexually Transmitted Disease, Shanghai Skin Disease Hospital,School of Medicine, Tongji University, Shanghai,200443, China
  1. Correspondence to Professor Pingyu Zhou, STD Institute, Shanghai Skin Diseases Hospital, Shanghai, Shanghai, China; zpyls{at}yahoo.com

Abstract

Objectives This study aimed to describe the clinical features of neurosyphilis in Chinese patients in an attempt to find clinical features that are helpful for the early identification of neurosyphilis.

Methods This retrospective study included people with syphilis who visited Shanghai Skin Disease Hospital between January 2010 and December 2020. Lumbar puncture was performed on those who met the inclusion and exclusion criteria. The diagnosis of neurosyphilis was based on clinical and laboratory findings. The parameters were analysed statistically.

Results Of the 3524 patients with neurosyphilis, 2111 (59.9%) and 1413 (40.1%) were asymptomatic and symptomatic neurosyphilis, respectively. General paresis was the most common type of symptomatic neurosyphilis (46.8%). The clinical manifestations of symptomatic neurosyphilis include psychiatric and neurotic symptoms, among which general paresis predominantly presented as psychiatric symptoms such as affective (66.7%) and memory disorder (72.9%). Tabes dorsalis is often presented as neurotic symptoms. One hundred fifty patients (10.6%) with symptomatic neurosyphilis presented candy signs, a rare and specific neurosyphilis symptom that is common in general paresis. Girdle sensation was presented in 13 patients, mainly with tabes dorsalis, which had not been reported in previous studies.

Conclusions Notably, the candy sign is identified as a specific symptom of general paresis, while girdle sensations are highlighted as a particular symptom of tabes dorsalis. This is the largest study describing the clinical spectrum of neurosyphilis since the onset of the penicillin era and could help doctors learn more about the disease. A comprehensive description of the possible clinical manifestations of late symptomatic neurosyphilis, particularly highlighting rare symptoms, can identify suspicious patients and prevent diagnostic delays.

  • SYPHILIS
  • NEUROLOGY
  • CHINA

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

What is already known on this topic

  • The clinical manifestations of neurosyphilis are varied and non-specific; to our knowledge, this is the largest study describing the clinical spectrum since the onset of the penicillin era.

What this study adds

  • This study found that general paresis and tabes dorsalis were the most common clinical types and mainly presented as psychiatric and neurotic symptoms, respectively. Notably, the proportion presenting candy signs was 10.6% and common in general paresis. Girdle sensation was presented in 13 patients, mainly with tabes dorsalis. Both symptoms have rarely been reported in previous studies.

How this study might affect research, practice or policy

  • Findings from this study help in knowing the clinical features of neurosyphilis and prompt diagnosis of neurosyphilis.

Introduction

Neurosyphilis was probably the most frequent type of dementia up to the first half of the 20th century.1 It affected a long list of artists, writers, musicians and philosophers.2 The discovery of penicillin led people to believe that syphilis is a night-extinct illness. However, syphilis is returning in most countries today, the incidence of syphilis has increased almost every year, increasing 52% during 2016–2020,3 and neurosyphilis is still a shader on earth.4 The annual incidence of neurosyphilis in various series has been reported to be 0.47 to 2.1 cases per 100 000 adults5 6 and in patients with clinical features of syphilis it is found that 3.5% of the patients had neurosyphilis.7

The neuropsychiatric manifestations of neurosyphilis can occur during any stage of Treponema pallidum infection.8 The treponeme disseminates throughout the nervous system within hours or days after primary infection8 and neurosyphilis may become either asymptomatic or symptomatic.7 Early neurosyphilis is usually asymptomatic. However, if left untreated, late neurosyphilis, which is mostly characterised by progressive dementia and psychiatric syndromes,9 can eventually cause irreversible damage to the central nervous system (CNS) and even become life-threatening,7 posing a significant public health disaster. Symptomatic neurosyphilis is classified into meningitis, meningovasculitis, general paresis, tabes dorsalis and gummatous syphilis based on clinical manifestations.7 Gummatous syphilis can be accompanied by neurological symptoms due to intracranial occupation, but generally does not present psychiatric symptoms in this study.

Most clinical descriptions of neurosyphilis originate from the pre-penicillin era.10 Recent studies on the clinical symptom description of neurosyphilis, which are mostly based on small samples, indicated that neurosyphilis exhibits a wide range of clinical manifestations and lacks specificity.11

We aimed to identify more comprehensive and specific clinical features suggestive of various types of neurosyphilis and to provide more clues for future clinical symptom identification.

Materials and methods

Study population and data collection

This retrospective study included people with syphilis who visited Shanghai Skin Disease Hospital between January 2010 and December 2020. We included patients with syphilis over the age of 18 years who had also undergone lumbar puncture (LP) screening. Exclusions were made for patients with only ocular syphilis, aural syphilis, cardiovascular syphilis and osteal syphilis without concomitant neurosyphilis. Additionally, we excluded pregnant or lactating patients, those with a family history of Alzheimer’s disease, LP failures or defective cerebrospinal fluid (CSF) (ie, bloody CSF, bacterial or fungal CSF), and patients who had received regular treatment for neurosyphilis before LP.

Clinical characteristics

All patients were managed with input from the hospital’s syphilis multidisciplinary team. For each patient, we collected a range of baseline information, including demographics, laboratory indexes, sexual orientation and comorbidities.

Psychiatric symptoms included disorders of consciousness, cognitive disorders, thought disorders, affective disorders, volitional behaviour disorders, memory disorders, intelligence disorders, insight, allolalia, dysarthria and dyssomnia. Neurotic symptoms included fever, dizziness, headache, nausea and vomiting, epileptic seizure, bowel and urine dysfunction, gait ataxia, paralysis, candy sign (orofacial dyskinesia), a sensation of walking on cotton, sensory disorders in the limbs, limb tremor, lightning-like pain, limb weakness and anaesthesia, girdle sensation (zonesthesia), hypotonia, lower limb oedema, a feeling walking on cobblestones and Charcot joints. In particular, girdle sensation only refers to the feeling of the chest or abdominal cavity, and the feelings of the lower limbs are not described by ‘girdle sensation’. Peripheral blood, CSF and other sexually transmitted diseases (STDs) tests were collected routinely from all patients. All laboratory tests were performed in the central laboratory of the Shanghai Skin Disease Hospital.

Diagnosis and classification of neurosyphilis

The diagnosis of neurosyphilis was based on clinical and laboratory findings. Peripheral venous blood was collected from all patients.

People with syphilis underwent LP in accordance with the Chinese national guidelines12 under the following conditions: (1) they presented with neurological or psychiatric symptoms; (2) they did not exhibit neurological or psychiatric symptoms, if: (a) non-treponemal test titres of patients experiencing treatment failure did not decrease by at least fourfold within 12 months after therapy, or their non-treponemal test titres shift from negative to positive or increased by more than fourfold compared with the latest treatment levels, and (b) the patient requested neurosyphilis evaluation due to persistent positive results on the non-treponemal test (ie, the test remains positive despite a sustained decline) after at least a year of anti-syphilitic treatment. Under the collaborative diagnostic assessment of multidisciplinary medical professionals, an LP was considered beneficial to uncover the progress of the disease.

Neurosyphilis was defined according to the following criteria, referencing both the Chinese national guidelines and the US/European guidelines:13–18(1) Reactive treponema and non-treponema tests in peripheral blood (screening for syphilis using rapid plasma reagin (RPR) test and confirmed by TPPA test), along with a reactive CSF-venereal disease research laboratory (CSF-VDRL) and CSF- TPPA tests; (2) when the CSF-VDRL test yielded non-reactive results, but a reactive CSF-TPPA test was present along with either or both of the following: (1) elevated CSF-protein (>50 mg/dL); (2) WBC count (>10 cells/µL) in the absence of other known causes of these abnormalities. The classification of symptomatic neurosyphilis was defined as the combination of clinical symptoms or signs consistent with neurosyphilis without other known causes for the clinical abnormalities and in accordance with the laboratory diagnostic criteria for neurosyphilis. Patients who suffered from more than two clinical types were defined as multiple types. Asymptomatic neurosyphilis was defined as conforming to the laboratory diagnosis of neurosyphilis without clinical symptoms or signs.

Statistical analysis

Continuous variables are presented as medians (IQR), whereas categorical variables are presented as n (%). Comparisons of baseline data were performed between patients with neurosyphilis and non-neurosyphilis. All continuous variables were tested for normality and compared by the non-parametric Mann-Whitney U-test to determine differences between symptomatic neurosyphilis and asymptomatic neurosyphilis. Categorical variables were compared using χ2 tests. A two-tailed p<0.05 was used to indicate statistically significant differences. All computations were performed using SPSS V.26.0 statistical software (IBM).

Results

Demographic characteristics

Between 2010 and 2020, a total of 71 484 cases of syphilis were referred to and/or diagnosed in our STD outpatient department. 3524 eligible participants (neurosyphilis) were included in this analysis. Among those who were diagnosed with neurosyphilis, 1413 (40.1%) patients were with symptomatic neurosyphilis and 2111 (59.9%) were with asymptomatic neurosyphilis, respectively. General paresis was the leading underlying symptomatic neurosyphilis (661, 46.8%) while other underlying symptomatic neurosyphilis, in descending order, were patients with multiple types (478, 33.8%), tabes dorsalis (174, 12.3%), meningovasculitis (81, 5.7%), gummatous syphilis (11, 0.8%) and meningeal syphilis (8, 0.6%) (figure 1). Among them, patients with multiple types were mainly characterised by general paresis combined with tabes dorsalis (370, 77.4%).

Figure 1

Flowchart showing the procedure used to recruit patients with neurosyphilis, multiple types (defined as symptomatic neurosyphilis patients who suffered from more than two clinical types). CSF, cerebrospinal fluid; LP, lumbar puncture; NS, neurosyphilis; VDRL, venereal disease research laboratory.

During the 11-year study period, there was a marked predominance of males (online supplemental figure 1A and table 1); the highest male-to-female ratio for patients with neurosyphilis occurred in 2011 (ratios: 3.4; 104 to 31). In terms of age, the neurosyphilis epidemic was concentrated among individuals who were 51–60 years of age. However, in the same age group of people with syphilis, the highest constituent ratio of patients with neurosyphilis was detected in those aged 61 years and above (online supplemental figure 1B and table 2). Among the patients with symptomatic neurosyphilis, the number with general paresis increased significantly (online supplemental figure 1C and table 3).

We identified 3524 (35.0%) patients with syphilis who were newly diagnosed with neurosyphilis during the 11-year study period (table 1). The group of 51–60 years (1238, 53.5%) had the highest concentration of patients with neurosyphilis (online supplemental table 2). Most patients with neurosyphilis were men (2546, 72.2%) and more than half of the patients (2242, 63.7%) had a serum RPR titre ≥1:32; 33.5% of patients with neurosyphilis had a history or co-infection of multiple STD infections (table 1).

Table 1

Baseline demographic and clinical characteristics of patients with neurosyphilis

Compared with patients with asymptomatic neurosyphilis, symptomatic neurosyphilis was more likely to be seen in the elderly (medians (IQR): 57 (50–62) years), men and individuals with higher CSF-VDRL. These patients were also more likely to be co-infected with multiple STDs and have diabetes, autoimmune disease or cancer (p<0.001). Conversely, they were less likely to be men who have sex with men (MSM) or people living with HIV (p<0.001; table 1).

The median time from onset of psychiatric and/or neurotic symptoms to patients seeking care was approximately 40 days (IQR 7.0–250.5 days); Significantly, it took another 211 days (IQR 98.0–399.0 days) before the diagnosis was confirmed. The time taken to diagnose meningovasculitis was the shortest (173.0 (61.0–309.5) days), while the time taken to diagnose gummatous was the longest (334.5 (198.5–478.8) days) (table 2).

Table 2

Delayed diagnosis of symptomatic neurosyphilis

Clinical features of symptomatic neurosyphilis

The clinical features of 1413 patients with symptomatic neurosyphilis are shown in table 3. The most common psychiatric symptoms were memory disorder (482, 72.9%) and affective disorder (441, 66.7%) in patients with general paresis. Candy signs (98, 14.8%) and limb tremors (40, 6.1%) were the most common neurotic symptoms of patients with general paresis. The number of cases of meningeal syphilis was small, only eight, but more than half of the patients had dizziness, headache, nausea and vomiting (5, 62.5%). Meningovasculitis was also associated with dizziness, headache, nausea and vomiting (40, 49.4%), along with limb weakness (42, 51.9%) and anaesthesia (23, 28.4%). The major symptoms of tabes dorsalis were limb anaesthesia (107, 61.5%), limb weakness (78, 44.8%), lightning-like pain (70, 40.2%), gait ataxia (48, 27.6%), walking on cotton feeling (43, 24.7%) and girdle sensation (10, 5.7%). Moreover, almost every patient with gummatous syphilis experienced dizziness, headache, nausea and vomiting (8, 72.7%). The neurological symptoms of multiple types were related to the clinical subtype. Overall, the most common psychiatric and neurotic symptoms were memory disorder (755, 53.4%) and limb weakness (279, 19.7%), respectively.

Table 3

Psychiatric and neurotic symptoms of 1413 patients with symptomatic neurosyphilis

Discussion

The recent surge in the reports detailing irreversible CNS injury caused by neurosyphilis is deeply concerning. Typically, symptomatic neurosyphilis is diagnosed during its advanced stages, with devastating consequences. As revealed by the findings of this study, the majority of patients were diagnosed in the late stage of neurosyphilis, including general paresis (661, 46.8%), multiple types (478, 33.8%) and tabes dorsalis (174, 12.3%). Urgent measures are imperative to mitigate the risk of adverse outcomes, underscoring the urgent need for early diagnosis of neurosyphilis.

Currently, the diagnosis of neurosyphilis primarily relies on a comprehensive evaluation of clinical characteristics, the laboratory examination of CSF and syphilis serological testing. However, the manifestations of neurosyphilis lack specificity, much like the skin manifestations of secondary syphilitic rash, earning it the moniker of a ‘great imitator’. Its clinical manifestations are diverse and can mimic various neurological and psychiatric diseases.1 Routine CSF testing has not been recommended for all patients,16 and the conventional CSF-VDRL, considered the ‘gold standard’ in the diagnosis of neurosyphilis, is often deemed insensitive.17 This poses challenges in accurately diagnosing neurosyphilis. The clinical symptoms of symptomatic neurosyphilis are varied and lack specificity, frequently resulting in misdiagnosis or delayed diagnosis,18 which can worsen the condition. Drawing from extensive data from a large sample of Chinese people, this study categorises psychiatric and neurotic symptoms in symptomatic neurosyphilis, providing a systematic description aimed at aiding clinical diagnosis and facilitating prompt treatment for patients.

To quantify the risk associated with delayed diagnosis of symptomatic neurosyphilis, we conducted an analysis of the duration between seeking care and receiving a final confirmed diagnosis. Our findings underscore the challenging nature of timely diagnosing symptomatic neurosyphilis. Non-specific clinical symptoms and a lack of clinician vigilance regarding neurosyphilis may contribute to prolonged diagnostic timelines. Among the various classifications of symptomatic neurosyphilis, meningovasculitis emerged as the condition diagnosed most quickly. This is likely due to its acute presentation often accompanied by cerebral infarction, making it more readily identifiable during hospitalisation. Conversely, other forms of symptomatic neurosyphilis typically took longer to diagnose, posing a heightened risk of diagnostic and treatment delays. Notably, a substantial portion of our study cohort comprised patients with general paresis and tabes dorsalis. Therefore, expedited diagnosis for individuals with these conditions holds significant clinical value. Consequently, our study focuses on analysing patients presenting with general paresis and tabes dorsalis.

Psychiatric and neurotic features of general paresis

Psychiatric symptoms are the main symptoms of patients with general paresis.9 Among these symptoms, memory disorder emerged as the most common psychiatric manifestation (482, 72.9%), consistent with findings from previous studies.11 Additionally, affective disorder was frequently observed in individuals with general paresis (441, 66.7%). Some studies have suggested that agitation could serve as a diagnostic indicator of general paresis.19 Furthermore, considering that memory or affective disorders are commonly associated with conditions such as Alzheimer’s disease, hypertension and diabetes, distinguishing between neurosyphilis and these diseases is of great significance.

Notably, our study revealed that candy sign was common in patients with general paresis (98, 14.8%) and multiple types (43, 9.0%). Coined in 1913 by Camillo Negro, the candy sign was described as pathognomonic of neurosyphilis.20 This dyskinesia is characterised by the patient appearing to suck on candy using their lower facial muscles, lips and jaw in a slow and repetitive manner.21 Despite its distinctiveness, there have been limited reports on the candy sign.21–23 In individuals with general paresis, dyskinesia can manifest, likely stemming from the extensive damage to the brain and spinal cord caused by syphilis. The presence of the candy sign may strongly suggest a diagnosis of neurosyphilis, especially in cases of late neurosyphilis. Therefore, when clinicians encounter patients with neuropsychiatric symptoms, especially candy signs, neurosyphilis should be kept in mind.

Psychiatric and neurotic features of tabes dorsalis

Tabes dorsalis was found in 12.3% of symptomatic neurosyphilis cases in this study. It is recognised that the symptoms of tabes dorsalis mainly include ataxic gait, lightning-like pains, impaired deep and proprioceptive sensation, and thermal sensation,7 due to the demyelination of the posterior roots of spinal nerves and posterior columns of the spinal cord. Among our patients with tabes dorsalis, predominant neurotic symptoms were observed, particularly limb anaesthesia (107, 61.5%), limb weakness (78, 44.8%), lightning-like pain (70, 40.2%), gait ataxia (48, 27.6%), walking on cotton feeling (43, 24.7%) and girdle sensation (10, 5.7%). Limb anaesthesia and lightning-like pain often manifest less frequently in the early stages of the disease and are often ignored by patients, indicating that once these symptoms appear, patients may be in the late stage. Notably, girdle sensation, indicative of spinal cord lesions,24 was also noted in tabes dorsalis, a finding not previously reported. The emergence of girdle sensation typically signifies a late stage of tabes dorsalis and is challenging to alleviate with conventional therapy.

Most clinical descriptions concerning neurosyphilis stem from the pre-penicillin era. Tabes dorsalis, characterised by chronic meningeal reaction caused by invasion of spirochete and destruction of adjacent neural tissue,7 was the most common manifestation of neurosyphilis in the pre-penicillin era.10 However, since the advent of the penicillin era, the spectrum of neurosyphilis has undergone significant changes, with reports indicating a marked decrease in the incidence of tabes dorsalis.9 25 In our cohort, tabes dorsalis accounted for 12.3% of cases. Conversely, general paresis remains the most common form of neurosyphilis, no matter whether in the pre-penicillin era10 or in the modern era.9 11 In our dataset, 46.8% of symptomatic neurosyphilis cases with general paresis. These findings suggest that while penicillin may inhibit the development of tabes dorsalis, its influence on general paresis development is minimal. Therefore, it is imperative to identify effective strategies to impede the progression of general paresis.

Real-world clinical studies indicate that some patients with early syphilis, even after standard treatment, may progress to symptomatic late-stage neurosyphilis.26 Our findings also suggest that neurosyphilis operates as a spectrum disease. As the disease progresses, its clinical spectrum can evolve, with asymptomatic neurosyphilis potentially developing into symptomatic neurosyphilis in very subtle ways, that may go unrecognised by the patients. Different types of neurosyphilis may share similar pathophysiological processes,11 leading to overlapping clinical manifestations. Therefore, it is crucial to monitor disease progression closely to accurately classify symptomatic neurosyphilis and ensure timely and appropriate intervention.

We conducted a comprehensive analysis of baseline data in patients with neurosyphilis and found that most were male and aged 50 years or older, consistent with our previous study.15 Additionally, male and elderly neurosyphilis patients were more likely to present with symptomatic than asymptomatic forms.27 Given that the cohort of patients diagnosed with syphilis at our institution between 2010 and 2020 had a gender ratio of 1:1, there was no gender bias in the present study. The impact of sex hormones on neurosyphilis accounts for this sex difference. Our previous studies indicated that the predisposition of elderly individuals might result from an impaired clearance of T. pallidum due to a weakened host immune response.15 Neurosyphilis has been associated with diabetes28 and hypertension29 in previous studies, which may also contribute to its higher prevalence among the elderly. In this study, diabetes was more likely to occur in patients with symptomatic neurosyphilis than in asymptomatic neurosyphilis ones, warranting further investigation into the underlying causes. The prevalence of symptomatic neurosyphilis was lower among people living with HIV and MSM in our study compared with Western countries.30 This difference might be attributed to regular annual syphilis screening and early treatment during the early or asymptomatic neurosyphilis stages of HIV infection in China.

There are several limitations to our study that need to be considered. First, as a retrospective study, it did not elucidate the underlying mechanisms. The association between clinical features and disease cannot infer causality because these clinical features are not universally present in symptomatic neurosyphilis. Second, we analysed the clinical features at a single baseline level, which prevented us from assessing the potential changes in the clinical syndrome over the course of disease progression. Third, our data were retrospectively analysed from a single institution, which may have introduced selection bias. Finally, the participants in our study were all Chinese; it is necessary to investigate patients from other biogeographical regions to confirm our findings. This limitation affects the generalisability of our conclusions to other races and ethnicities.

Conclusions

Using a database comprising 1413 Chinese patients with symptomatic neurosyphilis, we have conducted a detailed and comprehensive analysis of the manifestations and types of late-stage symptomatic neurosyphilis to aid in diagnosis. Our study revealed that, aside from meningovasculitis, various types of symptomatic neurosyphilis exhibited a high risk of diagnostic delay. General paresis and tabes dorsalis were the most prevalent types.

This is the largest study to describe the clinical spectrum of neurosyphilis since the advent of the penicillin era. By emphasising the significant diagnostic value of recognising symptoms such as the candy sign and girdle sensation, this research can help clinicians better understand the disease. A comprehensive description of the possible clinical manifestations of late-stage symptomatic neurosyphilis, particularly highlighting rare symptoms, can help identify suspicious cases and prevent diagnostic delays.

Abstract translation

This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by the Shanghai Skin Disease Hospital Ethics Committee (reference number: 2020-16). Written and informed consent was obtained from all patients. This study was reported according to STROBE. Participants gave informed consent to participate in the study before taking part.

References

Supplementary materials

Footnotes

  • YY and XG are joint first authors.

  • Handling editor Stefano Rusconi

  • YY and XG contributed equally.

  • Contributors YY, XG and PZ had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: YY, XG and PZ. Acquisition, analysis or interpretation of data: all authors. Drafting of the manuscript: YY, XG and PZ. Critical revision of the manuscript for important intellectual content: all authors. Statistical analysis: YY, XG. Administrative, technical or material support: XG and PZ. Supervision: PZ. PZ is responsible for the overall content as the guarantor.

  • Funding This work was supported by grants from National Natural Science Foundation of China (82072322, 82103739), Shanghai Science and Technology Commission (YDZX20193100002868, 17DZ2293300).

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.