Abstract
Aims
Toxic oil syndrome (TOS) was one of the first described forms of drug-induced pulmonary arterial hypertension (PAH). Its long-term clinical evolution remains poorly understood. The objectives of the present study were to provide new clinical, pathological and genetic insights into TOS-associated PAH (TOS-PAH), and to evaluate its long-term outcomes.
Methods
Patients diagnosed with TOS-PAH and included in the Spanish Registry of PAH (REHAP) were prospectively analysed.
Results
59 new cases were diagnosed between 1997 and 2025 (63% female; median age 47 years), including several in the past two decades. The median interval between TOS exposure and PAH diagnosis was 270.8 months (interquartile range (IQR) 204.9–398.0). Patients diagnosed in earlier periods were younger, with more advanced functional impairment and more severe haemodynamics. In contrast, recent cohorts showed a higher prevalence of cardiovascular and respiratory comorbidities. No significant differences were observed in overall or transplant-free survival across decades (p=0.677). Median transplantation-free survival was 97.2 months (IQR 64.2–199.4). Three patients achieved complete haemodynamic resolution. Genetic testing was negative in all evaluated patients. Pathological findings were comparable with those observed in other PAH forms, with some cases with significant venous remodelling.
Conclusions
TOS-PAH remains a distinct clinical entity, with new cases diagnosed decades after toxic exposure. Despite differences in presentation over time, its long-term clinical course appears similar to that of other PAH types.
Shareable abstract
PAH associated with exposure to toxic rapeseed oil represents a unique model of drug-induced PH in which the time from toxic exposure can be exceptionally long, with a relative benign course and complete PH resolution in some cases https://bit.ly/45Vr7Fb
Introduction
Toxic oil syndrome (TOS) is a multisystemic disease that occurred in Spain in 1981 due to massive intoxication as a result of the ingestion of denatured rapeseed oil with 2% aniline sold as cooking oil. This detonated a public scandal with epidemic proportions, mostly affecting the central and northwestern regions of the country, especially rural and suburban areas (figure 1) [1, 2]. In the acute phase of intoxication, patients typically present with nonproductive cough, dyspnoea, pleuritic chest pain, headache, fever and bilateral pulmonary infiltrates. A census of patients affected by TOS was created and supported by the Spanish government in 1981. In 1982, nearly 20 000 patients were reported, of whom at least 12 000 (approximately 60%) required hospitalisation and roughly 400 died [3, 4].
FIGURE 1.
Map of the Spanish regions showing a) the prevalence of toxic oil syndrome in the Spanish toxic oil syndrome census and b) the cases of pulmonary arterial hypertension diagnosed in patients from that census, as recorded in the Spanish Registry of Pulmonary Arterial Hypertension. Adapted from Posada de la Paz et al. [1].
Pulmonary hypertension (PH) was present in up to 20% of patients affected by TOS in the acute phase, many of whom were affected by acute lung injury and respiratory distress. The incidence of PH tended to decrease in patients affected by TOS from 3.5% in 1983 to 1.5% in 1985 [4]. Necropsy studies confirmed a non-necrotising endothelial disease with lesions consisting of intimal proliferation and medial hypertrophy of pulmonary vasculature [5]. Later, an association between greater degrees of pulmonary vascular lesions and longer follow-up was described, with evidence of plexiform lesions, endothelial injury, cell proliferation and perivascular inflammatory infiltrates [6], confirming the presence of an indistinguishable pathology compared with other forms of PAH [7]. While short-term mortality in patients with PAH in the first phase was extremely high, others developed PAH [6]. PAH associated with TOS is now recognised as a form of drug-induced PAH (DIPH) and is believed to have a similar clinical picture to other forms of PAH [8, 9]. Nevertheless, the chronic phase of PAH associated with TOS has never been described. We sought to analyse the long-term outcomes of patients with TOS included in the Spanish Registry of Pulmonary Arterial Hypertension (REHAP), as well as to evaluate novel genetical and pathological insights of these patients with a very long-term evolution of this form of DIPH.
Methods
Study subjects
The REHAP includes adult patients with group 1 or group 4 PH in Spain from 1998 [10]. All patients included in this study fulfilled the previous definition of PAH (mean pulmonary artery pressure (mPAP) ≥25 mmHg, pulmonary vascular resistance (PVR) ≥3 WU and pulmonary artery wedge pressure ≤15 mmHg) [9]. A Spanish census of patients affected by TOS (ORPHA:227972) was created in 1981. When patients were included in both registries, and after exclusion of other PH forms, they were classified as TOS-PAH. We included patients with TOS-PAH enrolled in REHAP up to 1 May 2025. Prevalent patients with PAH associated with TOS diagnosed before 1998 and alive at the time of registry initiation were included in this study. TOS-PAH patients who died before the creation of the registry in 1998 were not included in this work and are described extensively in a previous work detailing the acute phase of this disease [6]. The protocol was approved by the institutional review boards of all the participating hospitals. Patients provided written informed consent before being enrolled in the registry.
Variables
The clinical picture during the acute intoxication was obtained from the TOS census. All the variables related with PAH were collected from REHAP. These included demographic data, functional class (FC), 6-min walking distance (6MWD), cardiac biomarkers (brain natriuretic peptide (BNP) and the N-terminal portion of proBNP (NT-proBNP)), haemodynamic data (right atrial pressure (RAP), arterial oxygen saturation (SaO2), mixed venous oxygen saturation (SvO2), cardiac output (CO), cardiac index (CI), mPAP and PVR) and PAH-targeted treatments. The echocardiographic data included the right atrium (RA) area, pericardial effusion, tricuspid annular plane systolic excursion (TAPSE) and eccentricity index. Chest computed tomography (CT) images and pulmonary function tests at baseline were also reviewed.
Histology
Lung samples from two cases with TOS-PAH who underwent lung transplantation were available for this study. The samples were fixed in formalin and embedded in paraffin to obtain tissue blocks. Haematoxylin and eosin-stained images are shown in this work.
Genetics
Different next-generation sequencing panels were used across time in this cohort. Initially, a panel of 21 genes (HAP v.1.2) was used and 35 genes were covered (HAP v.3) on the basis of previous research data (supplementary table S1).
Statistical analysis
Continuous variables are expressed as median (interquartile range (IQR)) and were compared via the Mann–Whitney U-test. Categorical variables are expressed as n (%) and were compared using Fisher's exact test. Patients were followed up until the censoring date, 1 May 2025, or until the date of lung or heart–lung transplantation, or death. Patients lost to follow-up were censored at the date of their last recorded visit. Survival was assessed from the date of the first diagnostic right heart catheterisation using the Kaplan–Meier method. Both overall survival and transplant-free survival rates were estimated. To evaluate the potential impact of the time-period on outcomes, the cohort was divided by decade. A Mantel–Haenszel test was used to assess linear trends over time for categorical variables, using the first decade as the reference. A Jonckheere–Terpstra test with Kendall's ordinal correction was employed to evaluate trends in median values over time. Risk of death was evaluated using the 3-strata European model and the REVEAL 2.0 at baseline and using the 4-strata COMPERA model and the REVEAL 2.0 at first follow-up [9, 11]. Missing data were not imputed, and scores were calculated with the variables available. The global capacity of these stratification models was evaluated using the Harrel's C-index. All statistical analyses were performed via STATA v.14 (Stata Inc., TX, USA). Statistical significance was set at p<0.05.
RESULTS
Characteristics of patients with TOS-PAH
Since the establishment of the REHAP registry, 59 cases of TOS- PAH have been identified. According to the TOS census, during the acute phase of intoxication in 1981, 88.4% of these 59 cases presented with lung infiltrates, 83.7% with eosinophilia, 74.7% with myalgias, 48.8% with a generalised skin rash and 34.9% with sclerodermiform skin changes. The majority of patients were diagnosed within the Madrid metropolitan area (figure 1).
The median time from exposure to denatured rapeseed oil to the development of PAH was 270.8 months (IQR 204.9–398.0). Baseline characteristics of the cohort are summarised in table 1. 37 patients (62.7%) were female, with a median age of 46 years at diagnosis. The median 6-min walk distance (6MWD) was 405 m (IQR 313–477) and the median NT-proBNP level was 217 pg·mL−1 (IQR 95–381) in the 18 cases (30.5%) in which this biomarker was available. Transthoracic echocardiography at diagnosis showed mild right chamber dilation, with a median right atrial area of 22 cm2. Haemodynamic measurements revealed a median mPAP of 57 mmHg (IQR 44–74), a median PVR of 11.7 WU (IQR 8.3–17.8) and a mildly reduced median CI of 2.2 L·min−1·m−2 (IQR 1.9–2.6). Chest CT was reported as normal in the majority of cases, although radiological findings suggestive of emphysema were present in 11.8%, and signs of pulmonary veno-occlusive disease (PVOD) were observed in four cases (table 1).
TABLE 1.
Baseline characteristics of patients with toxic oil syndrome-associated pulmonary arterial hypertension
| Characteristic | Values |
|---|---|
| Age, years | 45.9 (36.8–57.4) |
| Sex, female | 37 (62.7) |
| Time from symptoms to diagnosis, months | 14.9 (4.5–60.8) |
| Heart failure signs or symptoms | 19 (32.0) |
| Syncope at diagnosis | 9 (15.3) |
| WHO functional class | |
| I | 1 (1.7) |
| II | 17 (28.8) |
| III | 34 (57.6) |
| IV | 7 (11.9) |
| Smoking history | |
| Active | 1 (1.7) |
| Never | 29 (49.2) |
| Former | 17 (28.8) |
| Unknown | 12 (20.3) |
| Cardiovascular risk factors at diagnosis | |
| Atrial fibrillation | 2 (3.4) |
| Systemic hypertension | 9 (15.3) |
| Coronary artery disease | 3 (5.1) |
| Diabetes | 3 (5.1) |
| Previously diagnosed respiratory disease | |
| OSA | 3 (5.1) |
| COPD | 3 (5.1) |
| Emphysema | 1 (1.7) |
| Asthma | 1 (1.7) |
| ILD | 1 (1.7) |
| Echocardiographic variables | |
| Pericardial effusion | 5 (8.5) |
| RV basal diameter, mm | 41 (36–46) |
| SPAP, mmHg | 91 (75–107) |
| LVEF, % | 65 (62–75) |
| RA area, cm2 | 22 (19–28) |
| TAPSE, mm | 19 (16–21) |
| Pulmonary function tests | |
| FVC, % | 91 (75–103) |
| FEV1, % | 85 (73–104) |
| FEV1/FVC | 75 (64–84) |
| DLCO, % | 65 (43–77) |
| TLC, % | 96 (81–102) |
| O2 sat at rest, % | 95 (92–96) |
| NT-proBNP, pg·mL−1 | 217 (95–381) |
| BNP, pg·mL−1 | 51 (40–415) |
| Distance 6MWT, m | 405 (313–477) |
| Haemodynamic values | |
| RAP, mmHg | 9 (5–13) |
| PAWP, mmHg | 10 (8–13) |
| mPAP, mmHg | 57 (44–74) |
| CO, L·min−1 | 3.8 (2.9–5.0) |
| CI, L·min−1·m−2 | 2.2 (1.9–2.6) |
| PVR, WU | 11.7 (8.3–17.8) |
| Sat O2, % | 64 (56–70) |
| Chest CT findings | |
| Emphysema | 4 (11.8) |
| Probable PVOD, ≥2 radiological signs | 4 (11.8) |
| Mosaic | 4 (11.8) |
| UIP pattern | 1 (2.9) |
| Other unspecific findings | 3 (8.8) |
| Normal | 18 (52.9) |
| Initial clinical vignette | |
| Pulmonary infiltrates | 38 (88.4) |
| Eosinophilia | 36 (83.7) |
| Myalgia | 33 (76.7) |
| Rash | 21 (48.8) |
| Sjögren syndrome | 12 (27.9) |
| Scleroderma-like syndrome | 15 (34.9) |
| Neuropathy | 26 (60.5) |
| Arthropathy | 18 (41.9) |
Data are presented as n (%) or median (interquartile range). OSA: obstructive sleep apnoea; ILD: interstitial lung disease; RV: right ventricle; SPAP: SPAP: systolic pulmonary artery pressure; LVEF: left ventricular ejection fraction; RA: right atrium; TAPSE: tricuspid annular plane systolic excursion; FVC: forced vital capacity; FEV1: forced expiratory volume in 1 s; DLCO: diffusing capacity of the lung for carbon monoxide; TLC: total lung capacity; NT-proBNP: N-terminal pro B-type natriuretic peptide; BNP: brain natriuretic peptide; 6MWT: 6-min walk test; RAP: right atrial pressure; PAWP: pulmonary artery wedge pressure; mPAP: mean pulmonary artery pressure; CO: cardiac output; CI: cardiac index; PVR: pulmonary vascular resistance; CT: computed tomography; PVOD: pulmonary veno-occlusive disease; UIP: usual interstitial pneumonia. CT signs suggestive of PVOD included ground-glass parenchymal opacities, septal lines and lymphadenopathy. A diagnosis of probable PVOD was considered when at least two of these three radiographical features were present in conjunction with a DLCO <50% of predicted.
Evolution over the last 40 years
13 cases were prevalent at the time of registry initiation; 12 diagnosed between 1980 and 1990 and one in the 1990s prior to registry creation. Demographic characteristics and disease severity at diagnosis evolved notably over time (table 2). A consistent female predominance was observed across decades. Patients diagnosed in earlier periods were significantly younger (median age 43.4 years in the 1980–1990 decade versus 81.6 years in the current decade; p<0.001 for trend). Additionally, individuals diagnosed in the 1980s and 1990s presented with more advanced functional impairment and more frequent heart failure at diagnosis (p=0.010 for trend). Conversely, systemic hypertension and smoking history became increasingly common in more recent cohorts (p=0.048 and p=0.010, respectively). Haemodynamic profiles were also more severe in earlier decades, with median mPAP values of 64.5 mmHg (IQR 50.0–74.5) in the 1980s and 68.0 mmHg (IQR 57.0–77.0) in the 1990s. In contrast, mPAP values decreased in more contemporary groups: 45.0 mmHg (IQR 43.0–65.0) in 2010–2020 and 44.0 mmHg (IQR 39.0–49.0) in the current decade. Overall, a decreasing trend in pulmonary pressures was observed over time (figure 2). At baseline, monotherapy was the most common treatment approach (62.7%), with dual oral combination therapy used in only 16.9% of patients. Systemic prostacyclin therapy was initiated in 30.5%, whereas upfront triple combination therapy was limited to two patients, both treated with inhaled iloprost. Notably, PAH resolved in three patients (5.1%), allowing withdrawal of vasodilators after a median of 175 months from diagnosis to the last right heart catheterisation, which showed normalised pulmonary pressures (supplementary table S2). In terms of outcomes, eight patients (13.6%) underwent lung or heart–lung transplantation and 27 (45.8%) died during follow-up. The median transplantation-free survival time was 97.2 months (IQR 64.2–199.4). The majority of deaths (77.8%) were due to heart failure or sudden cardiac death.
TABLE 2.
Comparison of baseline characteristics of patients diagnosed with toxic oil syndrome-associated pulmonary arterial hypertension over time
| Decade 1980–1990# (n=12) |
Decade 1990–2000 (n=13) |
Decade 2000–2010 (n=15) |
Decade 2010–2020 (n=17) |
Decade 2020– (n=2) |
p-value | |
|---|---|---|---|---|---|---|
| Sex, female | 10 (83.3) | 9 (69.2) | 7 (46.7) | 9 (52.9) | 2 (100.0) | 0.176 |
| Age at diagnosis, years | 43.4 (33.4–52.5) | 34.5 (24.5–38.5) | 47.3 (40.6–60.0) | 55.7 (42.6–68.8) | 81.6 (77.9–85.3) | <0.001 |
| Heart failure at diagnosis | 5 (45.5) | 9 (69.2) | 4 (26.7) | 1 (5.9) | 0 (0.0) | 0.002 |
| Syncope at diagnosis | 1 (9.1) | 5 (38.5) | 0 (0.0) | 3 (17.6) | 0 (0.0) | 0.643 |
| WHO FC at diagnosis, III/IV versus I/II | 10 (83.3) | 12 (92.3) | 11 (73.3) | 6 (35.3) | 2 (100.0) | 0.010 |
| Smoking history, current or former versus never-smoker | 0 (0.0) | 2 (28.6) | 5 (41.7) | 10 (62.3) | 1 (50.0) | 0.002 |
| Atrial fibrillation at diagnosis | 0 (0.0) | 0 (0.0) | 1 (14.3) | 1 (7.7) | 0 (0.0) | 0.686 |
| Systemic hypertension at diagnosis | 0 (0.0) | 1 (50.0) | 1 (14.3) | 5 (38.5) | 2 (100.0) | 0.048 |
| Coronary artery disease | 0 (0.0) | 0 (0.0) | 1 (14.2) | 2 (15.4) | 0 (0.0) | 0.440 |
| Diabetes | 0 (0.0) | 1 (50.0) | 0 (0.0) | 2 (16.7) | 0 (0.0) | 0.864 |
| Pulmonary function tests | No observations | |||||
| FVC, % | 78.8 (62.5–88.0) | 81.1 (77.0–98.7) | 92.0 (75.0–99.0) | 103.5 (90.9–112.5) | 0.002 | |
| FEV1, % | 75.2 (69.0–105.5) | 82.0 (74.0–93.5) | 79.0 (73.0–99.0) | 99.0 (85.0–107.0) | 0.028 | |
| FEV1/FVC | 85.0 | 109.0 | 60.0 | 74.0 (64.0–82.0) | 0.524 | |
| DLCO, % | 70.5 (53.5–82.0) | 61.0 (42.5–73.0) | 65.5 (36.0–84.0) | 65.0 (47.0–76.0) | 0.614 | |
| NT-proBNP, pg·mL−1 | No observations | No observations | 117.0 (51.0–140.0) | 281.0 (95.0–381.0) | 2065.0 | 0.090 |
| BNP, pg·mL−1 | No observations | 30.0 | No observations | 51.0 (50.0–778.0) | No observations | |
| Distance 6MWT, m | 340.0 (120.0–406–0) | 390.0 (240.0–477.0) | 400.0 (317.0–495.0) | 450.0 (410.0–528) | 103.0 | 0.020 |
| Haemodynamic values | ||||||
| RAP, mmHg | 6.5 (4.0–17.5) | 11.0 (6.0–13.0) | 8.0 (5.0–10.0) | 7.0 (5.0–12.0) | 10.5 (8.0–13.0) | 0.765 |
| PAWP, mmHg | 8.5 (5.5–11.5) | 9.0 (8.0–14.0) | 9.5 (5.0–12.0) | 11.0 (9.0–13.0) | 15.0 (13.0–17.0) | 0.129 |
| mPAP, mmHg | 64.5 (50.0–74.5) | 68.0 (57.0–77.0) | 52.0 (41.0–77.0) | 45.0 (43.0–65.0) | 44.0 (39.0–49.0) | 0.012 |
| CO, L·min−1 | 3.2 (2.5–4.8) | 3.2 (2.6–4.1) | 4.7 (3.7–5.6) | 3.8 (3.1–5.2) | 3.7 (3.1–4.4) | 0.125 |
| CI, L·min−1·m−2 | 2.1 (1.5–2.5) | 2.0 (1.7–2.4) | 2.6 (2.1–3.5) | 2.5 (2.1–2.6) | 2.3 (2.1–2.4) | 0.130 |
| PVR, WU | 13.0 (9.9–21.5) | 18.9 (11.5–30.3) | 9.8 (7.3–14.4) | 9.6 (8.3–13.6) | 7.8 (7.5–8.0) | 0.012 |
| Sat O2, % | 60.0 (44.0–70.0) | 56.5 (55.0–58.0) | 63.0 (60.0–71.0) | 69.5 (66.5–71.5) | 68.0 | 0.011 |
Data are presented as n (%) or median (interquartile range). FC: functional class; FVC: forced vital capacity; FEV1: forced expiratory volume in 1 s; DLCO: diffusing capacity of the lung for carbon monoxide; NT-proBNP: N-terminal pro B-type natriuretic peptide; BNP: brain natriuretic peptide; 6MWT: 6-min walk test; RAP: right atrial pressure; PAWP: pulmonary artery wedge pressure; mPAP: mean pulmonary artery pressure; CO: cardiac output; CI: cardiac index; PVR: pulmonary vascular resistance. A Mantel–Haenszel test was used to test the linear trend over time in relation to the reference decade for categorical variables, using only valid cases for trend comparisons. A Jonckheere–Terpstraa test with Kendall ordinal correction was used to test the linear trend between different medians over time. #: As noted in the methods, patients diagnosed before the creation of the Spanish Registry of Pulmonary Arterial Hypertension in 1998 and alive at the time of the registry initiation were included in this study. Bold p-values indicate statistical significance.
FIGURE 2.
Evolution of haemodynamic parameters in patients with pulmonary arterial hypertension associated with toxic oil syndrome. a) Mean pulmonary artery pressure (mPAP); b) cardiac output; c) pulmonary vascular resistance (PVR). RHC: right heart catheterisation.
According to the European Society of Cardiology (ESC)/European Respiratory Society (ERS) 3-strata model, most patients were initially classified as intermediate risk (22.0% low, 67.8% intermediate and 10.2% high). Risk reassessment using the COMPERA 4-strata model at first follow-up demonstrated substantial improvement: 42.0% of patients were categorised as low risk, 30.0% as intermediate–low, 24.0% as intermediate–high and only 4.0% as high risk (figure 3a,b). Similarly, REVEAL v.2.0 risk stratification showed progressive improvement (figure 3c,d). Both risk models demonstrated prognostic utility in this population. The Harrell's C-index for the ESC/ERS model was 0.66 at baseline and 0.73 at first follow-up; for the REVEAL v.2.0 model, it was 0.69 at both time points (figure 4). Both models also accurately predicted survival free from lung and heart–lung transplantation (supplementary figure S1). Notably, no significant differences in overall survival or transplantation-free survival were observed across different decades of diagnosis (log-rank test; p=0.677; supplementary figure S2). Patients with features of PVOD demonstrated a nonsignificant trend for worse survival free of lung transplantation (median survival of 66.2 versus 105.3 months; IQR 64.2–68.2 versus 64.5–199.4; p=0.258).
FIGURE 3.
Evolution of risk stratification a) according to the European 3-strata model at baseline and the COMPERA 4-strata model at first follow-up and b) according to the REVEAL 2.0 score, at baseline and at first follow-up. The COMPERA 4-strata risk was calculated using the available variables at first re-evaluation, which included N-terminal pro-B-type natriuretic peptide or B-type natriuretic peptide only in 14 cases. In the remaining cases, the score was calculated when the other two variables (functional class and distance walked in the 6-min walk test) were available.
FIGURE 4.
Observed overall survival rates in patients diagnosed with pulmonary arterial hypertension associated with toxic oil syndrome in the Spanish Registry of Pulmonary Arterial Hypertension (REHAP). a) Overall survival based on the European 3-strata score at baseline; b) Overall survival based on the COMPERA 4-strata model at first follow-up; c) Overall survival based on the REVEAL 2.0 score at baseline; and d) at first follow-up. The COMPERA 4-strata risk was calculated using the available variables at first re-evaluation, which included N-terminal pro-B-type natriuretic peptide or B-type natriuretic peptide only in 14 cases. In the remaining cases, the score was calculated when the other two variables (functional class and distance walked in the 6-min walk test) were available. ESC: European Society of Cardiology; ERS: European Respiratory Society.
Genetic analyses
During follow-up, 19 patients (32.2% of the cohort) underwent genetic analysis. Nevertheless, the genetic results were negative in all patients. The genes analysed and the methodology are shown in supplementary table S1.
Pathology
In this work, we present the images of two patients with TOS-PAH. The first patient was a woman diagnosed in June 2003 after 21 years of the intoxication, and was transplanted in October 2020, 17 years after PAH diagnosis (patient 1 in supplementary table S2). This patient had a baseline low risk, never smoked, did not have any comorbidity, had normal pulmonary function tests (forced expiratory volume in 1 s (FEV1) 83% and forced vital capacity (FVC) 82% of predicted values) and a diffusing capacity of the lung for carbon monoxide (DLCO) of 63% at diagnosis. She was treated with sequential triple therapy until lung transplantation in 2020. Chest CT before transplantation did not demonstrate any signs of emphysema nor other parenchymal abnormalities (figure 5a–c). In this patient, advanced precapillary arterial remodelling was present, with unspecific inflammatory infiltrates, moderate signs of emphysema and without evidence of significant venular involvement (figure 6a–c). The second patient is a case diagnosed of PAH in 2019, 38 years after the intoxication (patient 53 in supplementary table S2). This man had systemic hypertension and was a former smoker of 10 pack-years. In low risk at baseline, tadalafil was initiated in monotherapy, but was subsequently treated with a triple systemic vasodilator. The pulmonary function tests in this case were unremarkable (FVC of 105%, FEV1 107% and total lung capacity of 89% of predicted values), except for the presence of a severely decreased DLCO of 47%. The chest CT demonstrated mild paraseptal emphysema, mediastinal lymphadenopathies and septal thickening, signs of probable PVOD (figure 5d–f). The pathology was compatible with this latter diagnosis, demonstrating pigmented intra-alveolar macrophages and venular remodelling, among other signs of predominant venous remodelling (figure 6d–f).
FIGURE 5.
Chest computed tomography (CT) images of patients diagnosed with pulmonary arterial hypertension (PAH) associated with toxic oil syndrome who underwent lung transplantation during follow-up. a–c) Images correspond to a female patient transplanted 17 years after PAH diagnosis, showing no radiological signs of emphysema or other parenchymal lung disease. d–f) Images correspond to a male patient with a history of smoking, transplanted 6 years after PAH diagnosis. The chest CT showed mild paraseptal emphysema, mediastinal lymphadenopathy and septal thickening, which are findings suggestive of probable pulmonary veno-occlusive disease.
FIGURE 6.
Histopathological images from patients diagnosed with pulmonary arterial hypertension (PAH) associated with toxic oil syndrome who underwent lung transplantation during follow-up. a–c) Images (×10 view, haematoxylin and eosin (H&E) stain) correspond to a female patient transplanted 17 years after PAH diagnosis, showing emphysematous changes (asterisk), typical arteriolar remodelling (black arrows) and plexiform lesions (green arrow). d–f) Images (200 μm H&E stain) correspond to a male patient with a history of smoking, transplanted 6 years after PAH diagnosis. In addition to typical arteriolar remodelling (black arrow), these samples also showed widespread venous remodelling (red arrow), emphysematous changes (asterisks) and inflammatory infiltrates (orange arrow).
Discussion
This is the first study to examine the long-term evolution of PAH associated with exposure to toxic rapeseed oil. We observed a remarkably high incidence of PAH in the TOS cohort, characterised by an unusually long latency period between toxin exposure and disease manifestation. Interestingly, we observed a wide phenotypic variability, including PVOD-like and typical arteriolar forms, and a few cases that experienced a complete haemodynamic resolution.
Combining previously reported cases (n=40) [6] with the 59 newly identified in this study, a total of 99 TOS-PAH cases have been documented among an estimated 20 000 individuals affected by the toxic oil. This represents an exceptionally high prevalence of PAH in this unique population. Contrary to initial expectations, new cases have continued to emerge over the decades following the original intoxication event, with diagnoses spanning from the 1980s to the present. However, determining the true incidence of PAH among individuals exposed to TOS remains challenging. Furthermore, comparisons with the incidence of idiopathic PAH in the general population are limited by underreporting and regional disparities in diagnostic vigilance. While most cases in recent decades have been diagnosed in the Madrid metropolitan area, the initial outbreak affected several regions across central Spain. The underdiagnosis of TOS-PAH in other areas remains a plausible concern (figure 1). A striking example of potential underdiagnosis is that the two cases identified in the current decade had a median age at diagnosis of 81.6 years.
The median latency between toxin exposure and PAH diagnosis was noteworthy at 270.8 months, far exceeding the timelines reported for other toxin-induced forms of PAH, such as dasatinib-associated PAH (median 40 months) [12] or fenfluramine-induced PAH (up to 66 months) [13]. Several patients who exhibited systemic features during the acute phase of intoxication (e.g. sclerodermiform skin lesions) continued to display these signs chronically, which may have delayed PAH diagnosis by leading to misattribution of dyspnoea to other causes. The prolonged latency period in TOS-PAH underscores the uniqueness of this model within the spectrum of DIPH and toxin-induced PH, as similar long-term manifestations have only been described in rare conditions such as PAH associated with pre-tricuspid congenital cardiac septal defects [14].
Over the past four decades, there has been a discernible trend toward milder clinical presentation at the time of diagnosis. Patients identified in recent years exhibited better functional class, longer 6MWDs and less severe haemodynamic compromise. This likely reflects increased awareness in the TOS population, and detection of milder disease forms. Although patients with TOS-PAH initially faced limited treatment options, their overall clinical course was relatively favourable, showing a response to pulmonary vasodilators comparable with that of idiopathic PAH, as reflected in risk score improvements over time. More than 5% of patients experienced full resolution of PAH, and the median transplantation-free survival exceeded 8 years. These findings align with outcomes reported in other DIPH subtypes such as dasatinib-associated PAH, in which long-term vasodilator withdrawal has occasionally been possible [15]. In contrast, methamphetamine-associated PAH (meth-PAH) has been associated with poorer survival outcomes [16].
From a genetic standpoint, we did not identify PAH-associated variants in TOS-PAH patients, differing from findings in other DIPH cases where the toxin may act as a “second hit” in genetically susceptible individuals [17]. For example, BMPR2 mutations have been linked to earlier disease onset in fenfluramine-induced PAH [18]. Environmental and genetic factors may influence susceptibility to PAH in DIPH. Previous studies have described associations between systemic sclerosis and dasatinib exposure [19], and higher rates of smoking in meth-PAH patients compared with idiopathic PAH [20].
Histologically, early TOS studies revealed generalised non-necrotising vasculopathy affecting also the pulmonary circulation [2]. In the current study, we report two TOS-PAH cases with long-standing disease and characteristic findings; one with classic plexiform vasculopathy and another with significant venous remodelling. The latter has previously been associated with radiographical features suggestive of PVOD and low DLCO [21]. Mild emphysema was also noted in both cases, echoing the pathological findings in other forms of PAH, in which emphysema can be observed. On one hand, certain forms of heritable PAH not associated with significant respiratory conditions, such as TBX4- or FLNA-related PAH, often present with some degree of emphysema, both radiologically and pathologically [22, 23]. On the other hand, hypoxia-inducible factor may play a role in this phenomenon, potentially contributing to parenchymal changes under chronic hypoxic conditions [24]. The progression of disease in TOS-PAH, from an acute systemic vascular disease to a chronic, localised pulmonary vascular disorder, provides a rare and valuable window into the natural history of DIPH. This represents a unique clinicopathological continuum that underscores the long-term consequences of environmental toxin exposure and offers a compelling opportunity for learning from a preventable public health disaster. Unfortunately, the precise mechanisms underlying why some individuals develop PAH and others do not have yet to be fully elucidated.
Limitations
This study has several limitations, primarily due to its retrospective design. A clear time-period bias exists, as diagnostic and therapeutic approaches varied significantly across decades, potentially influencing clinical characteristics, treatment strategies and outcomes. Furthermore, the inclusion of long-term survivors (patients with persistent but moderate disease who survived the early post-exposure period) introduces a significant survivorship bias. These individuals may not represent the full clinical spectrum of TOS-PAH, particularly the most severe early cases. Nonetheless, our data indicate that overall prognosis has remained relatively stable over time. Additionally, as with many voluntary registries such as REHAP, selection and survivorship biases are inherent, given that the most severe PAH cases are often underrepresented [25].
Conclusions
TOS-PAH represents a unique model of DIPH, marked by an exceptionally long latency between toxin exposure and the clinical diagnosis of PAH. Some patients showed prominent venous remodelling, whereas others experienced full haemodynamic resolution. No PAH-predisposing genetic variants were identified. The distinct natural history and pathology of TOS-PAH provide valuable insights into toxin-induced vascular remodelling and the long-term impact of environmental exposures.
Acknowledgments
We recognise the crucial contribution of M. Posada de la Paz from the Instituto de Salud Carlos III, Spanish Government, who dedicated almost his entire career investigating TOS. We also recognise the important role of M. Ángel Gómez Sánchez (previous coordinator of the Pulmonary Hypertension Unit, Hospital Universitario 12 de Octubre, Madrid, Spain) as the first person who described TOS-PAH, and we especially dedicate this work to all affected patients with this terrible disease. We express our gratitude to Merck Sharp & Dohme and Ferrer for supporting REHAP with an unrestricted educational grant. We gratefully acknowledge all investigators who form part of this registry. We also thank the Registry Coordinating Centre, S&H Medical Science Service, for their quality control, logistic and administrative support and especially Salvador Ortiz, PhD, Universidad Autónoma de Madrid and Statistical Advisor S&H Medical Science, for the statistical analysis of the data presented in this paper. The authors declare that no artificial intelligence applications have been used for this work.
Footnotes
Provenance: Submitted article, peer reviewed.
This article has an editorial commentary: https://doi.org/10.1183/23120541.01091-2025
Coordinators of REHAP: Pilar Escribano-Subias (Spain) and Isabel Blanco (Spain). Investigators of the REHAP Group: I. Alarcón de la Lastra Cubiles, Hospitales Universitarios Vall d'Hebron – Sant Pau, Barcelona; S. Alcolea, Hospital Universitario La Paz, Madrid; M. Álvarez, Hospital Universitario Ramón y Cajal, Madrid; Á. Aurtenetxe Pérez, Hospital Universitario Basurto, Bilbao; M.J. Azpiroz Franch, Hospitales Universitarios Vall d'Hebron – Sant Pau, Barcelona; J.A. Barbera, Hospital Clínic i Provincial de Barcelona, Barcelona; I. Blanco, Hospital Clínic i Provincial de Barcelona, Barcelona; S. Cadenas Menéndez, Hospital Universitario de Salamanca (Clínico), Salamanca; J. Carrillo, Hospital Universitario Rey Juan Carlos, Madrid; C.I. Chamorro Fernández, Hospital Virgen de los Lirios, Alicante; J.M. Cifrián, Hospital Universitario Marqués de Valdecilla, Cantabria; T. Clavero Sánchez, Hospital Universitario de Salamanca (Clínico), Salamanca; M.J. Del Cerro, Hospital Universitario Ramón y Cajal, Madrid; R. Del Pozo Rivas, Hospital Universitario Juan Ramón Jiménez, Huelva; F. Díaz Chantar, Hospital Universitario Juan Ramón Jiménez, Huelva; J.A. Domingo Morera, Hospital Universitario Miguel Servet, Zaragoza; L. Dos Subirá, Hospitales Universitarios Vall d'Hebron - Sant Pau, Barcelona; P. Escribano-Subías, Hospital Universitario 12 de Octubre, Madrid; F.J. García Hernández, Hospital Universitario Virgen del Rocío, Seville; E. Garrido-Lestache Rodríguez-Monte, Hospital Universitario Ramón y Cajal, Madrid; S. Gomara de la Cal, Hospital Universitario Miguel Servet, Zaragoza; F. Gonçalves Dos Santos Carvalho, Hospital Universitari Germans Trias i Pujol, Barcelona; R. González León, Hospital Universitario Virgen del Rocío, Seville; F.J. Guerra Ramos, Complejo Hospitalario Universitario Insular-Materno Infantil, Las Palmas; M. Lázaro Salvador, Hospital Universitario de Toledo, Toledo; M. López-Meseguer, Hospital Universitario Vall d′Hebrón, Barcelona; R. López Reyes, Hospital Universitario y Politécnico La Fe, Valencia; A. Martínez-Meñaca, Hospital Universitario Marqués de Valdecilla, Cantabria; T. Mombiela Ramírez de Ganuza, Hospital Universitario Gregorio Marañón, Madrid; V.M. Mora Cuesta, Hospital Universitario Marqués de Valdecilla, Cantabria; J. Morgado, Hospital Universitario Juan Ramón Jiménez, Huelva; M. Noris Mora, Hospital Universitario Son Espases, Islas Baleares; N. Ochoa Parra, Hospital Universitario 12 de Octubre, Madrid; J. Ribas Sola, Hospital Universitari de Bellvitge, Barcelona; S. Rodríguez Suárez, Hospital Universitario Virgen del Rocío, Seville; J. Rueda Soriano, Hospital Universitario y Politécnico La Fe, Valencia; B. Sáez Giménez, Hospital Universitario Vall d′Hebrón, Barcelona; E. Sala Llinas, Hospital Universitario Son Espases, Islas Baleares; H. Sintes Permanyer, Hospital Universitari Germans Trias i Pujol, Barcelona; C. Soto Abánades, Hospital Universitario La Paz, Madrid; J.A. Tenes, Hospital Universitario Ramón y Cajal, Madrid; G. Torroba Balmori, Hospital de Dénia Marina Salud, Alicante.
Author contributions: Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work: A. Cruz-Utrilla, A. Cantero-Acedo, A. Tenes and P. Escribano-Subias. Drafting the work or revising it critically for important intellectual content: A. Cruz-Utrilla, A. Cantero-Acedo, A. Tenes, A.B. Enguita, J.F. Delgado-Jiménez, M. López-Meseguer, A. Martínez-Meñaca, M. Lázaro Salvador, I. Martín de Miguel, E. Gutiérrez Ortiz, J. Segovia-Cubero, J.A. Tenorio, I. Blanco and P. Escribano-Subias. Final approval of the version to be published: A. Cruz-Utrilla, A. Cantero-Acedo, A. Tenes, A.B. Enguita, J.F. Delgado-Jiménez, M. López-Meseguer, A. Martínez-Meñaca, M. Lázaro Salvador, I. Martín de Miguel, E. Gutiérrez Ortiz, J. Segovia-Cubero, CSV, I. Blanco and P. Escribano-Subias. Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved: A. Cruz-Utrilla, A. Cantero-Acedo, A. Tenes, A.B. Enguita, J.F. Delgado-Jiménez, M. López-Meseguer, A. Martínez-Meñaca, M. Lázaro Salvador, I. Martín de Miguel, E. Gutiérrez Ortiz, J. Segovia-Cubero, CSV, I. Blanco and P. Escribano-Subias.
Ethics statement: The protocol was approved by the institutional review boards of all the participating hospitals. Patients provided written informed consent before being enrolled in the registry.
Conflicts of interests: All authors have confirmed that they have no conflicts of interest to declare.
Support statement: A. Cruz-Utrilla holds a research contract Juan Rodes from the Instituto de Salud Carlos III, Ministerio de Ciencia, Innovación y Universidades, Spanish Government (JR23/00071). I. Martín de Miguel holds a research contract Río Hortega from the Instituto de Salud Carlos III, Ministerio de Ciencia, Innovación y Universidades, Spanish Government (CM23/00235). P. Escribano-Subias holds a research grant I+D+I from the Instituto de Salud Carlos III, Ministerio de Ciencia, Innovación y Universidadades, Spanish Government (PI21/01690). Funding information for this article has been deposited with the Open Funder Registry.
Supplementary material
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary material
00720-2025.SUPPLEMENT
Data availability
The data that support the findings of this study are available on request from the corresponding author, A. Cruz-Utrilla. The data are not publicly available due to restrictions (data containing information that could compromise the privacy of research participants).
References
- 1.Posada de la Paz M, Philen RM, Borda AI. Toxic oil syndrome: the perspective after 20 years. Epidemiol Rev 2001; 23: 231–247. doi: 10.1093/oxfordjournals.epirev.a000804 [DOI] [PubMed] [Google Scholar]
- 2.Manuel Tabuenca J. Toxic-allergic syndrome caused by ingestion of rapeseed oil denatured with aniline. Lancet 1981; 318: 567–568. doi: 10.1016/S0140-6736(81)90949-1 [DOI] [PubMed] [Google Scholar]
- 3.Kilbourne EM, Rigau-Perez J G, Heath CW, et al. Clinical epidemiology of toxic-oil syndrome: manifestations of a new illness. N Engl J Med 1983; 309: 1408–1414. doi: 10.1056/NEJM198312083092302 [DOI] [PubMed] [Google Scholar]
- 4.Gómez-Sánchez MA, de Juan MJ M, Gómez-Pajuelo C, et al. Pulmonary hypertension due to toxic oil syndrome. A clinicopathologic study. Chest 1989; 95: 325–331. doi: 10.1378/chest.95.2.325 [DOI] [PubMed] [Google Scholar]
- 5.Martinez-Tello FJ, Navas-Palacios JJ, Ricoy JR, et al. Pathology of a new toxic syndrome caused by ingestion of adulterated oil in Spain. Virchows Arch A Path Anat and Histol 1982; 397: 261–285. doi: 10.1007/BF00496569 [DOI] [PubMed] [Google Scholar]
- 6.Gomez-Sanchez MA, De La Calzada CS, Gomez-Pajuelo C, et al. Clinical and pathologic manifestations of pulmonary vascular disease in the toxic oil syndrome. J Am Coll Cardiol 1991; 18: 1539–1545. doi: 10.1016/0735-1097(91)90688-6 [DOI] [PubMed] [Google Scholar]
- 7.Guignabert C, Aman J, Bonnet S, et al. Pathology and pathobiology of pulmonary hypertension: current insights and future directions. Eur Respir J 2024; 64: 2401095. doi: 10.1183/13993003.01095-2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kovacs G, Bartolome S, Denton CP, et al. Definition, classification and diagnosis of pulmonary hypertension. Eur Respir J 2024; 64: 2401324. doi: 10.1183/13993003.01324-2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Humbert M, Kovacs G, Hoeper MM, et al. 2022 ESC/ERS Guidelines for the diagnosis and treatment of pulmonary hypertension. Eur Heart J 2022; 43: 3618–3731. doi: 10.1093/eurheartj/ehac237 [DOI] [PubMed] [Google Scholar]
- 10.Escribano-Subias P, Blanco I, López-Meseguer M, et al. Survival in pulmonary hypertension in Spain: insights from the Spanish registry. Eur Respir J 2012; 40: 596–603. doi: 10.1183/09031936.00101211 [DOI] [PubMed] [Google Scholar]
- 11.Benza RL, Kanwar MK, Raina A, et al. Development and validation of an abridged version of the REVEAL 2.0 risk score calculator, REVEAL Lite 2, for use in patients with pulmonary arterial hypertension. Chest 2021; 159: 337–346. doi: 10.1016/j.chest.2020.08.2069 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Shah NP, Wallis N, Farber HW, et al. Clinical features of pulmonary arterial hypertension in patients receiving dasatinib. Am J Hematol 2015; 90: 1060–1064. doi: 10.1002/ajh.24174 [DOI] [PubMed] [Google Scholar]
- 13.Souza R, Humbert M, Sztrymf B, et al. Pulmonary arterial hypertension associated with fenfluramine exposure: report of 109 cases. Eur Respir J 2008; 31: 343–348. doi: 10.1183/09031936.00104807 [DOI] [PubMed] [Google Scholar]
- 14.Van der Feen DE, Bartelds B, de Boer RA, et al. Assessment of reversibility in pulmonary arterial hypertension and congenital heart disease. Heart 2019; 105: 276–282. doi: 10.1136/heartjnl-2018-314025 [DOI] [PubMed] [Google Scholar]
- 15.Weatherald J, Chaumais MC, Savale L, et al. Long-term outcomes of dasatinib-induced pulmonary arterial hypertension: a population-based study. Eur Respir J 2017; 50: 1700217. doi: 10.1183/13993003.00217-2017 [DOI] [PubMed] [Google Scholar]
- 16.Zamanian RT, Hedlin H, Greuenwald P, et al. Features and outcomes of methamphetamine-associated pulmonary arterial hypertension. Am J Respir Crit Care Med 2018; 197: 788–800. doi: 10.1164/rccm.201705-0943OC [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Humbert M, Deng Z, Simonneau G, et al. BMPR2 germline mutations in pulmonary hypertension associated with fenfluramine derivatives. Eur Respir J 2002; 20: 518–523. doi: 10.1183/09031936.02.01762002 [DOI] [PubMed] [Google Scholar]
- 18.Humbert M, Deng Z, Simonneau G, et al. BMPR2 germline mutations in pulmonary hypertension associated with fenfluramine derivatives. Eur Respir J 2002; 20: 518–523. doi: 10.1183/09031936.02.01762002 [DOI] [PubMed] [Google Scholar]
- 19.Toya T, Nagatomo Y, Kagami K, et al. Dasatinib-induced pulmonary arterial hypertension complicated with scleroderma: a case report. Eur Heart J Case Rep 2019; 3: ytz025. doi: 10.1093/ehjcr/ytz025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kolaitis NA, Zamanian RT, De Jesus Perez VA, et al. Clinical differences and outcomes between methamphetamine-associated and idiopathic pulmonary arterial hypertension in the pulmonary hypertension association registry. Ann ATS 2021; 18: 613–622. doi: 10.1513/AnnalsATS.202007-774OC [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Barrios E, Gómez-Sanchez MA, Velázquez M, et al. Absence of acute vasoreactivity response in patients with pulmonary arterial hypertension associated with toxic oil syndrome. Eur Heart J 2013; 34: Suppl. 1, P321. doi: 10.1093/eurheartj/eht307.P321 [DOI] [Google Scholar]
- 22.Prapa M, Lago-Docampo M, Swietlik EM, et al. First genotype-phenotype study in TBX4 syndrome: gain-of-function mutations causative for lung disease. Am J Respir Crit Care Med 2022; 206: 1522–1533. doi: 10.1164/rccm.202203-0485OC [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Stourm L, Grynblat J, Savale L, et al. Pulmonary hypertension in patients carrying FLNA loss-of-function variants. Eur Respir J 2025; 65: 2401132. doi: 10.1183/13993003.01132-2024 [DOI] [PubMed] [Google Scholar]
- 24.Myronenko O, Foris V, Crnkovic S, et al. Endotyping COPD: hypoxia-inducible factor-2 as a molecular “switch” between the vascular and airway phenotypes? Eur Respir Rev 2023; 32: 220173. doi: 10.1183/16000617.0173-2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Miller DP, Gomberg-Maitland M, Humbert M. Survivor bias and risk assessment. Eur Respir J 2012; 40: 530–532. doi: 10.1183/09031936.00094112 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary material
00720-2025.SUPPLEMENT
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author, A. Cruz-Utrilla. The data are not publicly available due to restrictions (data containing information that could compromise the privacy of research participants).