Shrinking lung syndrome in systemic lupus erythematosus: a systematic review

Shrinking Lung Syndrome (SLS) is a rare pulmonary complication of Systemic Lupus Erythematosus (SLE) and other connective tissue diseases. SLS is characterised by dyspnoea, restrictive lung pattern and elevated hemidiaphragms. The pathophysiology of SLS is unclear and there are currently no defined diagnostic criteria or standardised treatment guidelines. Increasing clinical awareness is essential to reduce diagnostic delay and improve patient outcomes. A literature search was conducted on the MEDLINE and Embase databases from January 1946 to 29th December 2025 to identify studies evaluating shrinking lung syndrome in patients with systemic lupus erythematosus. Data extracted included age, sex, presenting symptoms, imaging, PFTs, phrenic nerve or diaphragm studies, immunologic profile, treatment and follow-up outcomes. A total of 34 studies were included, all of which consisted of case reports, case series and one cross-sectional study. We pooled 120 patients, 86% of which were females and ages ranged from 12 to 67 years, including paediatric cases. Dyspnoea was present in 98% of patients. Elevated hemidiaphragms were seen on imaging in 78% of patients. Pulmonary Function Tests (PFTs) showed a restrictive ventilatory pattern in 91% of patients. Corticosteroids were the first-line treatment in 93% of patients. Symptomatic improvement was seen in 77% of patients. Reported follow-up periods ranged from 3 weeks to 20 years. Early recognition of SLS and tailored immunosuppressive therapy can achieve good outcomes for patients.

Systemic Lupus Erythematosus (SLE) is a chronic multisystem autoimmune disease, which is often associated with significant morbidity and mortality. The global incidence of SLE is estimated to be 5.14 per 100,000 person-years (1). Women of childbearing age are most commonly affected by SLE, with women being ten times more likely to develop SLE than men. SLE is less commonly reported in paediatric and elderly patient populations. The prevalence of SLE is highest amongst Black, Hispanic and Asian populations, with earlier onset and higher severity found in Black populations. Although the etiology of SLE remains unknown, there are several known genetic, endocrine, environmental factors that are associated with the development of the disease. The morbidity and mortality of SLE are largely determined by major organ involvement, including musculoskeletal, renal, pulmonary, cardiovascular and gastrointestinal systems (2). Patients with SLE have a 2.6 higher risk of mortality compared to the general population, highlighting the importance of early diagnosis and treatment (3).

Pulmonary involvement has been reported in 20%–90% of patients with SLE, with this wide variation reflecting differences in diagnostic criteria, disease severity, symptom reporting and study populations. Pulmonary involvement in patients with SLE is associated with lower survival rates than those without any pleuropulmonary manifestations (82.2% vs. 95.6%, p = 0.030). Some patients develop permanent lung damage after SLE diagnosis (7.6% at 5 years, 11.6% at 10 years) and 4%–5% of cases have pulmonary manifestations as their presenting symptom (4). Types of pulmonary involvement in SLE include lupus pleuritis, pleural effusion, acute lupus pneumonitis, shrinking lung syndrome, interstitial lung disease, diffuse alveolar hemorrhage (DAH), pulmonary arterial hypertension, and pulmonary embolism. Pleuritis and pleural effusion are the most common types of pulmonary involvement, found in up to 45%–60% of patients (4).

Shrinking Lung Syndrome (SLS) was first described in 1965 by Hoffbrand and Beck, as a rare pulmonary complication of systemic autoimmune disease, most commonly seen in SLE but sometimes found in Sjogren's syndrome, scleroderma, and rheumatoid arthritis. SLS is estimated to affect 0.5%–1.53% of all patients with SLE. It is associated with dyspnea and pleuritic chest pain, alongside reduced lung volumes and absence of parenchymal disease. Typical investigation findings include elevated hemidiaphragms on chest x-ray and restrictive pattern on pulmonary function tests (PFTs). There is currently no defined diagnostic criteria for SLS, and diagnosis require exclusion of a variety of illnesses which cause pleuritic pain or dyspnoea, including pneumonia, pleurisy, pericarditis, pulmonary embolism, interstitial lung disease, chronic obstructive pulmonary disease, heart failure, pulmonary hypertension and musculoskeletal chest wall disorders (5). Variability in diagnostic criteria and overlap of symptoms with other pulmonary complications of SLE make diagnosis of SLS challenging and can result in delayed and missed diagnosis.

The pathophysiology of shrinking lung syndrome is uncertain and there are many hypotheses looking to explain its unique clinical features. Proposed mechanisms include diaphragmatic dysfunction, phrenic nerve paralysis, pleural inflammation and myositis (5).

There are no standardised treatment guidelines for SLS. Common treatments reported involve corticosteroids, beta-2 agonists, theophylline, analgesia, and chest physiotherapy. Rituximab is also an option for steroid-refractory cases and azathioprine as a steroid-sparing option. Immunosuppressants such as cyclophosphamide, azathioprine, mycophenolate mofetil, and methotrexate are usually not successful if steroids fail. Response to treatment is variable and whilst most patients achieve some clinical improvement, complete recovery is uncommon (5).

The current literature on shrinking lung syndrome is predominantly case reports and small case series. The aim of this narrative review is to synthesise the available literature on shrinking lung syndrome in patients with systemic lupus erythematosus, focusing on clinical presentation, diagnostic features, proposed pathophysiological mechanisms, management strategies and outcomes.

Study design

A literature review was performed to identify published studies describing shrinking lung syndrome in patients with systemic lupus erythematosus. The research question was structured using the PICO framework. The population (P) was made up of patients with confirmed diagnosis of systematic lupus erythematosus (SLE); the exposure (I) was the development of shrinking lung syndrome; there was no comparator (C) applicable, as this review is descriptive in nature; and the outcomes (O) included clinical features, investigation findings, management plans and follow-up outcomes.

Search strategy and data sources

An electronic search was conducted on the MEDLINE and Embase databases from January 1946 to 29 th December 2025 to identify studies evaluating shrinking lung syndrome in patients with systemic lupus erythematosus. The search terms included were “systemic lupus erythematosus” or “SLE”, with “shrinking lung syndrome”. These terms were searched within article titles and abstracts where appropriate.

Study selection and data extraction

Studies were screened and data extracted by one reviewer (NB), with a second reviewer (DP) independently verifying study selection. All articles were initially screened for relevance by title and abstract. Review articles, case reports and case series published in English and involving human subjects were considered for inclusion. Case reports and clinical studies were included if they reported on patient demographics, clinical presentation, timing of SLS relative to SLE diagnosis, imaging findings, management strategies and outcomes of shrinking lung syndrome in patients with systemic lupus erythematosus. Full-texts were reviewed when available, however studies for which full texts could not be accessed were excluded, as adequate data extraction and critical appraisal could not be done based solely on abstracts.

Data analysis

Relevant data were extracted from included studies and synthesised narratively. Studies were eligible for synthesis if they reported original patient data. Where sufficient data were available, outcomes were summarised using counts and percentages. No meta-analysis was performed due to the heterogenous nature of the included studies and the predominance of case reports and case series. Ethical approval was not required as this study involved only analysis of previously published data.

Study selection and characteristics

The literature search identified 107 records across MEDLINE (n = 34) and Embase (n = 73). From MEDLINE, 21 studies were included after screening. From Embase, 13 records remained after removal of duplicates and screening. This led to a total of 34 included studies, as shown in Table 1 and Figure 1. The included studies were heterogeneous in design, consisting of 19 case reports, 9 case series, 5 case series with literature reviews and 1 cross-sectional cohort study (39). Most studies were retrospective, with varying levels of detail regarding patient demographics, clinical features, and follow-up duration. Across the 33 studies reporting original patient data, the total pooled number of patients was 120. Some studies provided individual patient-level data, whereas others reported in group summaries. The studies were conducted across multiple geographic regions, reflecting a broad, though predominantly adult, patient population.

Demographics

The demographic profile of patients with shrinking lung syndrome reflected the typical pattern observed in systemic lupus erythematosus, as summarised in Table 1. Patient ages ranged from 12 to 67 years, with the majority being adults. The mean age across studies reporting original data was approximately 33 years. Sex was reported for 118 of the 120 pooled patients, of which 102 (86%) were female and 16 (14%) were male. Notable outliers included the youngest paediatric patients and older adults up to 67 years, though these were uncommon. Overall, the age and sex distribution mirrors that of the general SLE population.

Clinical presentation

Shrinking lung syndrome was the initial manifestation of SLE in 22% of patients, whereas the majority of patients developed SLS after an established diagnosis of SLE. Among the patients with available data, the median interval from SLE diagnosis to SLS onset was 48 months, with a wide range from 0 weeks (concurrent onset) to 38 years. Clinical presentations varied in onset, with some patients developing symptoms gradually over months to years, such as one case describing progressive dyspnoea over one year prior to diagnosis (7). On the other hand, other patients experienced a more abrupt onset, including a case of acute SLS following a two month period of medication non-compliance and prednisolone withdrawal (33). The difference in presentations highlights the variability of progression in this disease.

Dyspnoea was the main clinical presentation of shrinking lung syndrome, reported in 115 of 117 patients with data available (98%). Pleuritic chest pain was also common, affecting 85 of 101 patients with data available (81%). A subset of patients also developed orthopnoea, particularly in those with more severe diaphragmatic involvement. Fever, fatigue, arthralgia, and cough were also frequently described, though inconsistently quantified across studies. The severity of symptoms varied across cases. Some patients experienced features of acute respiratory distress, including rapid or shallow breathing, tachypnoea, orthopnoea and paradoxical abdominal motion. Overall, these findings are consistent with the multicentre literature review (19), which looked at data for 89 patients. The review found that dyspnoea was reported in 99% of patients, pleuritic chest pain in 76%, fever in 19% of patients, dry cough in 8% and orthopnoea occasionally.

Imaging findings

Chest imaging from the pooled 120 patients demonstrated characteristic features of shrinking lung syndrome. Chest x-rays were performed in 105 patients (88%) and CT scans were performed for 81 patients (68%). Elevated hemidiaphragms were identified in 93 patients (78%). Reduced lung volumes were observed in 52 patients (43%) and basal or laminar atelectasis was identified in 41 patients (34%). Pleural effusions or pleural thickening were noted in 37 patients (31%). Early or baseline imaging were occasionally normal, for example, in one case, the patient's baseline chest x-ray was normal, and serial chest x-rays showed gradual bilateral elevation of hemidiaphragms (20). In a case series of seven patients, two had normal chest x-rays and three had normal CT scans (36). Pulmonary parenchymal architecture was generally preserved, with no evidence of significant parenchymal fibrosis or diffuse interstitial changes reported in the 120 pooled patients. Diaphragmatic fluoroscopy or ultrasound were done in some patients, showing findings that were not seen on standard chest x-ray and CT scan. These extra tests confirmed reduced diaphragmatic excursion in 19 patients and no diaphragmatic movement in one patient (33). In one patient, dynamic MRI found that both diaphragms and all auxiliary respiratory muscles moved normally during forced ventilatory ventilation (21). Overall, imaging demonstrated a consistent pattern of elevated hemidiaphragms, reduced lung volumes, and basal or laminar atelectasis, with generally preserved lung parenchyma and minimal pleural involvement.

Pulmonary function tests

Pulmonary function test (PFT) data were available for 117 patients. A restrictive ventilatory pattern was identified in 107 of 117 patients (91%), while six patients demonstrated an obstructive pattern, one having documented marked bronchodilator reversibility. Initial forced vital capacity (FVC) values ranged from 15% to 99% predicted, with most patients demonstrating moderate to severe restriction. Total lung capacity (TLC) was similarly reduced, with reported values ranging from 32% to 88%. Diffusing capacity for carbon monoxide (DLCO) was reduced in the majority of patients when reported, with values ranging from 20% to 96% predicted. Two patients had preserved DLCO (≥ 80% predicted). The transfer coefficient (KCO) was reported in 33 patients. KCO was preserved or elevated in 90% of patients, with values ranging from 106% to 200% predicted in most patients. Reduced KCO was observed in 5 patients with values of 37%, 47%, 54%, 59% and 73% predicted.

Follow-up PFT data demonstrated partial improvement in several patients after treatment. In one case series of nine patients, two patients showed improvement in restrictive pattern, six remained stable, and one demonstrated complete resolution (19). Many individual cases also showed increases in FVC and TLC over follow-up, however, persistent restrictive patterns were common despite clinical improvement. Overall, PFT findings were characterised by a predominantly restrictive pattern with reduced lung volumes, frequently reduced DLCO, and preserved or increased KCO.

Phrenic nerve studies

Phrenic nerve conduction studies were reported in four patients within the pooled cohort. In one patient, phrenic nerve potentials were reported as normal (7). In another patient, magnetic phrenic nerve stimulation showed normal diaphragm twitch pressures before and after treatment, indicating normal contractility (38). In a different patient, surface diaphragm EMG was conducted. On the left diaphragm, EMGdi response to cervical magnetic stimulation (CMS) was abolished, indicating absent diaphragmatic electrical activity, and on the right diaphragm, EMGdi response was delayed, suggesting partial impairment. Mouth pressure twitch (Pm,t) was 1 cm H2O (normal > 10 cm H2O) and abdominal circumference decreased, indicating that the diaphragm was mechanically ineffective. After corticosteroid treatment, EMGdi responses improved on both sides, abdominal movement normalised and Pm,t increased slightly, indicating some diaphragm mechanical activity was restored (37). These different findings show variability in phrenic nerve conduction studies results.

Laboratory findings

Reporting of laboratory data varied across studies. Antinuclear antibody (ANA) results were available in 82 of 120 pooled patients, and 81 (99%) were ANA positive. Anti-double stranded-DNA (anti-dsDNA) results were reported in 96 patients, of whom 78 (81%) were positive. Anti-Ro antibody results were available in 39 patients and were positive in 25 (64%). Complement levels (C3 and/or C4) were reported in 49 patients and 34 (69%) had reduced complement levels. In some cases, increased consumption of complements were described during active SLE flares (25). Lymphopenia and raised inflammatory markers (high CRP and/or ESR) were reported in a subset of patients, however they were inconsistently documented across studies.

Treatment

The treatment of shrinking lung syndrome was dominated by corticosteroids as the first line, with 112 out of 120 pooled patients (93%) being given corticosteroids. Corticosteroids were administered orally, typically with doses ranging from 15–60 mg/day of prednisolone, or intravenously as high-dose methylprednisolone pulses. Both oral prednisolone and IV methylprednisolone were given to 15 patients. Conventional immunosuppressants were used frequently either alongside corticosteroids or as escalation treatment. Azathioprine was used in 33 patients (28%), IV or oral cyclophosphamide was administered in 23 patients (19%), mycophenolate mofetil in 21 patients (18%) and methotrexate in 10 patients (8%). Rituximab was used in 12 patients (10%) and Belimumab in 2 patients (2%), both typically used in refractory cases. Theophylline was used in 12 patients (10%). In one patient, treatment with theophylline was associated with the development of paroxysmal supraventricular tachycardia and atrial flutter, which persisted after treatment discontinuation and required long-term treatment with verapamil (13). Supportive therapies were reported in a subset of cases, including inhaled beta-2-agonists in 19 (16%) patients and pulmonary physiotherapy in 4 patients (3%). Overall, corticosteroids were consistently used as first-line treatment for SLS, with conventional immunosuppressants and rituximab given to patients with persistent or refractory disease.

Outcomes

Symptomatic or clinical outcomes were reported in 104 patients: improvement was observed in 80 patients (77%), while 18 patients (17%) remained stable, and 6 patients (6%) showed no improvement or worsened. Follow-up PFT data was recorded in 85 patients: partial recovery of pulmonary function was demonstrated in 54 patients (64%), stabilisation of the restrictive ventilatory pattern was observed in 19 patients (22%) and complete normalisation in 6 patients (7%). Persistent restrictive patterns with no improvement were noted in 6 patients (7%). Follow-up duration ranged from 3 weeks to 20 years. One death was reported across the cohort as a result of SLS.

Assessment of risk of bias

The included studies consisted predominantly of case reports and small case series, which is expected given the rarity of shrinking lung syndrome. Risk of bias was assessed by one reviewer (NB) using the Joanna Briggs Institute (JBI) critical appraisal tool. Critical appraisal demonstrated that the majority of included studies showed low risk of bias, supporting the reliability of the synthesized findings.

Although shrinking lung syndrome in systemic lupus erythematosus remains uncommon, the analysis of 120 pooled patients demonstrates reproducible patterns across demographics, clinical presentation, treatment response and outcomes. This discussion will contextualise these findings within the existing literature, with a particular focus on proposed pathogenesis, diagnostic challenges and treatment response.

In this pooled cohort, SLS predominantly affected females, which is consistent with the epidemiology of SLE, as 90% of patients with SLE are women in most studies (4). The age distribution was broad, ranging from adolescence to late adulthood, indicating the SLS is not confined to adult populations. Dyspnoea was reported in the majority of patients at presentation and was frequently accompanied by pleuritic chest pain, consistent with the characteristic symptom profile of SLS described previously (17). SLS occurred both in patients with established SLE and as an early or initial manifestation of disease, with isolated cases being reported in the context of undifferentiated connective tissue disease (17). Several reports highlighted the substantial functional limitation experienced by patients despite often subtle investigation findings, including cases in which patients developed marked orthopnoea and severe restriction on pulmonary function tests despite minimal parenchymal changes on imaging. Although the duration of symptoms prior to diagnosis was not consistently reported in the pooled cases, prior literature has highlighted diagnostic delay as a frequent challenge. In a previous review, 48% of patients experienced persistent symptoms for more than one year before SLS was recognised (19). The nonspecific nature of dyspnea, preserved lung parenchyma on imaging and the need to exclude more common cardiopulmonary complications of SLE likely contribute to delayed diagnoses.

Chest imaging demonstrated a consistent pattern of elevated hemidiaphragms in the majority of cases, frequently accompanied by reduced lung volumes and basal or laminar atelectasis, reflecting reduced diaphragmatic excursion. Pleural effusions or pleural thickening were also observed in a minority of patients, which may reflect the idea that pleuritic inflammation can cause diaphragmatic dysfunction in SLS (21). However, inconsistent reporting of pleural findings across studies means this mechanism remains partially substantiated. Lung parenchyma was generally preserved, with no evidence of interstitial lung disease, distinguishing SLS from primary interstitial lung disease as a cause of restriction. In some cases, baseline chest x-ray was reported as normal, with subsequent imaging revealing progressive diaphragmatic elevation, highlighting the dynamic nature of SLS. Fluoroscopy and ultrasound were occasionally performed, demonstrating either reduced or absent diaphragmatic movement. These findings align with previous reports that suggest diaphragm weakness in SLS is functional rather than structural. Pleural inflammation and reflex inhibition, likely caused by pain or irritation, may be the cause of diaphragmatic dysfunction, rather than intrinsic diaphragm damage (21).

Dynamic MRI in a single reported patient showed normal movement of both diaphragms and all auxiliary respiratory muscles, suggesting that diaphragmatic dysfunction may not always explain lung restriction in patients with SLE. However, this finding is only from one case, so the use of dynamic MRI as a diagnostic tool in SLS remains unclear and requires evaluation in a larger population (21).

The pattern of pulmonary function test results – disproportionate reductions in lung volumes relative to preserved efficiency of gas transfer (KCO) – points to an extrapulmonary cause of restriction. A reduced DCLO with normal or elevated KCO therefore serves as a useful diagnostic clue. The preservation of KCO despite significantly reduced FVC and TLC is consistent with prior literature suggesting that alveolar-capillary gas exchange remains intact, distinguishing SLS from interstitial lung disease. Patients often reported symptomatic relief before improvements on PFTs were seen, suggesting that the restrictive pattern was due to functional diaphragmatic suppression rather than permanent lung or muscle damage. Overall, PFTs quantify the severity of restriction and indicate that diaphragmatic dysfunction plays a mechanistic role in SLS.

Non-volitional testing, such as phrenic nerve stimulation, often demonstrated preserved diaphragm contractility. A previous literature review found that phrenic nerve conduction studies were normal in 76% of patients, with abnormalities only observed in a minority of cases (19). Normal phrenic nerve conduction studies in the majority of patients does not suggest that neuropathy is contributing to decreased lung volumes, rather supports a functional limitation. These observations suggest that reduced lung volumes in SLS are a result of reflex inhibition of the diaphragm, potentially mediated by pain or pleural inflammation. This mechanism may explain why symptoms can improve rapidly with immunosuppressive therapy and why PFT improvement may lag behind symptomatic improvement (31).

The EMGdi results from one patient showed absent activity on the left and delayed activity on the right, consistent with a demyelinating lesion of the phrenic nerves, pointing to a neuropathic cause rather than a primary muscle problem. The improvement in EMGdi responses following corticosteroid therapy further supports an immune-mediated neuropathy. This case contrasts with most other reports in the literature, where SLE patients with diaphragmatic weakness typically have normal phrenic nerve conduction, highlighting that this case is a rare example of neuropathic diaphragm involvement (37). Two further cases point to a myopathic rather than neuropathic mechanism: muscle biopsy demonstrated an endomysial T-lymphocytic infiltrate in one case (29), while another described a CD20+ B-cell-predominant myositis that responded to mycophenolate mofetil (11).

Laboratory findings mainly reflected the expected antibody profile of systemic lupus erythematosus rather than a distinct serologic pattern specific to SLS. ANA, anti-dsDNA and anti-Ro positivity alongside complement consumption were frequently documented, which could suggest that some cases of SLS occurred during periods of increased SLE activity. However, reporting of laboratory data was inconsistent across studies, limiting the ability to identify strong or consistent antibody associations. Therefore, there is currently no specific laboratory finding which could be used to predict the development of SLS. SLS has also been described in Sjogren's syndrome, scleroderma, and rheumatoid arthritis, indicating that its underlying mechanism is unlikely to be exclusive to SLE. The reasons for its stronger reported association with SLE specifically remain unclear and warrant further investigation.

There are currently no standardised treatment guidelines for SLS. Corticosteroids are used as the first-line treatment in the majority of patients. Immunosuppressants, such as azathioprine, mycophenolate mofetil, cyclophosphamide and methotrexate, were sometimes used as steroid-sparing therapy or in addition to corticosteroids if initial response was incomplete. Rituximab and Belimumab were reported to be used in refractory or relapsing cases of SLS. Several cases had a significant improvement in symptoms and sustained clinical improvement following anti-CD20 treatment. Previous reviews have suggested that B cell depletion may improve SLS by suppressing autoantibody production and immune activation, although the exact mechanisms are unknown (31). Theophylline was used in a minority of cases as an adjunct, though it carries a documented risk of cardiac arrythmia that requires monitoring. Supportive therapies, including inhaled beta-2 agonists and pulmonary physiotherapy, were occasionally used but rarely as the primary management. A structured inspiratory muscle training program has also been reported as a promising non-pharmacological treatment, improving respiratory muscle strength and diaphragm thickness on ultrasound even in patients with preserved spirometry (8). It is unclear what their exact role in recovery is more broadly, as there is no objective, isolated data available comparing outcomes before and after most other supportive therapies. As there are no controlled studies, the treatment of SLS currently relies primarily on creating an individualised management plan according to clinical and symptomatic response.

The majority of patients experienced improvement in their symptoms after treatment, often before improvement was seen on PFTs, with partial rather than complete PFT recovery being more common. Worsening of restrictive patterns on PFTs after treatment was uncommon. The gap between symptomatic and physiological recovery reinforces that SLS is a reversible, functional process rather than fixed structural lung damage. Follow-up periods in the reports varied and therefore limits conclusions about the long-term implications of the disease. Overall prognosis appears to be good with appropriate treatment and severe long-term impact on function is uncommon. Mortality directly related to SLS is rare but not absent. One death attributed to SLS was recorded (16), alongside three unrelated deaths in the same cohort, indicating that while overall prognosis remains favorable, fatal outcomes can occur in severe or refractory disease.

Limitations

The available literature currently consists mainly of case reports and small case series, which limits the generalisability of findings. Selection bias is possible, as unusual or severe cases are more likely to be reported. Clinical features, investigations and treatments were not reported uniformly in all studies. Many investigations and serology testing were not performed or reported in all patients. In particular, specialised investigations such as phrenic nerve studies, EMG or dynamic MRI were not performed in most patients, so underlying pathophysiology remains hypothetical. Inconsistent reporting of the doses, durations, and modes of administration of corticosteroids, immunosuppressants and biologics, makes it difficult to draw firm conclusions regarding the most effective treatment options. Similarly, limited data on treatment-specific outcomes prevent direct comparisons between individual treatment approaches. Variability in duration and thoroughness of follow-ups across studies makes it difficult to assess the long-term outcomes of SLS. Prognosis specifically among patients with refractory or treatment-resistant SLS could not be assessed, as long-term follow-up data for this subgroup was inconsistently reported across the included studies. This heterogeneity highlights both how rare shrinking lung syndrome is and the largely observational nature of the evidence base. This review was limited to database search of MEDLINE and Embase, and no review protocol was prospectively registered.

Shrinking lung syndrome is a rare but important pulmonary manifestation of systemic lupus erythematosus. This review highlights the consistent pattern of dysponea, restrictive pulmonary function and preserved lung parenchyma observed across reported cases. Diagnosis of SLS remains challenging due to the absence of standardised diagnostic criteria and the non-specific nature of many clinical and imaging findings. Therefore, the diagnosis of SLS is frequently delayed as exclusion of other causes of restrictive lung disease in patients with SLE is necessary. Corticosteroids are the mainstay of treatment, with addition of immunosuppressants and biologics in selected patients. The overall prognosis seems to be favourable, with most patients experiencing symptomatic improvement or stabilisation after treatment. Further studies are required to clarify the underlying mechanisms of SLS, to establish standardised diagnostic and management approaches and to understand the long-term implications of the disease.