The Deleterious Consequences of a Nonsteroidal Anti-inflammatory Drug in an Eisenmenger Patient

We report on a 36-year-old patient known for Eisenmenger's syndrome due to a non-restrictive ventricular septal defect (VSD) with predominant right-left shunt associated with class I pulmonary arterial hypertension (PAH). The patient was under chronic macitentan (10mg od) and sildenafil (20mg tid) treatment. Because of neck pain due to muscular contraction for one week, she ingested a total of 9g of paracetamol over 3 days (1g every 8 hours), and one blister pack of 10 tablets of ibuprofen 400mg, (nonsteroidal anti-inflammatory drug, NSAID) i.e., a total dose of 4g over 2 days. She additionally took 300 mg of tolperisone (a muscle relaxant drug) once daily. As she subsequently developed over two days a rapidly progressive dyspnea, she presented to the emergency department (ED) of a local hospital. On arrival at the ED, the patient presented with high blood pressure of 169/110 mmHg, accompanied by desaturation requiring an oxygen therapy of 15 L/min (FIO2 at 60%) to reach an oxygen saturation at rest of 80% (usual level at room air at 80-83%). On physical examination, the patient was tachypneic (20 breaths/min), had a normal pulmonary auscultation and showed no clinical signs of right or left heart failure. Because of the unstable clinical condition, the patient was transferred to the tertiary center. The blood gas analysis at arrival indicated hypoxemic respiratory failure without elevation of lactate. There were no clinical or laboratory signs for an infectious or an inflammatory etiology. COVID-19 and Influenza B + A PCR nasopharyngeal swab were negative. Laboratory tests revealed an increased NT-proBNP. Hemoglobin and platelet count were within usual ranges (Table). Computed tomography ruled out pulmonary embolism, edema and atelectasis. Further laboratory investigations eliminated common alternative causes of secondary hypertension including normal thyroid hormones, plasma metanephrines, and aldosterone. The patient was admitted to the intermediate care unit. A transthoracic echocardiography (TTE) showed the previously known anatomy with a small left ventricle with normal systolic function, a dilated and hypertrophied right ventricle with normal systolic function and the large non-restrictive perimembranous VSD with a right-left shunt. The patient’s condition deteriorated rapidly, requiring a transfer to the intensive care unit (ICU) for high-flow oxygen therapy for 24 hours (Table). In parallel, the platelet count decreased to a minimum value of 22 G/L (Table). Based on the clinical picture of concomitant desaturation, arterial hypertension and thrombocytopenia, we concluded that NSAID toxicity was likely responsible for an increase of PVR and worsening of the right-left shunt. We therefore increased the sildenafil dosage to 40 mg tid and maintained macitentan 10 mg od. The patient’s condition improved over the following 9 days, with normalization blood pressure, recovery of platelet count and return of oxygen saturation to usual levels (Table) .Table 1Clinical and laboratory parametersDaySystolic Blood PressureDiastolic Blood pressureO2 Saturation - SaO2/SpO2FiO2ThrombocytesCreatinineHemoglobinNt-pro BNP(mmHg)(mmHg)(%)(%)(G/L)(μmol/L)(G/L)(ng/L)2 years1309580AA1137122290-6 months1179080AA11580227108-6.5 months80AA11286221-3 months1258580AA8386227-2 months80AA1148021701691117580%60872201015115911167100%-2161968230%422193153998335%37652164157927932%252075150868228%22622126137918324%255921471291008328%38622248125817824%4721991298980AA63223111101228980AA8579217Abbreviations: fraction of inspired oxygen = FiO2, NT-proBNP = N-terminal pro-B-type natriuretic peptide), AA =ambient air Open table in a new tab Abbreviations: fraction of inspired oxygen = FiO2, NT-proBNP = N-terminal pro-B-type natriuretic peptide), AA =ambient air We describe three adverse side effects of NSAIDs in a patient with Eisenmenger’s syndrome.1)Arterial Hypertension: Prostaglandin E2 (PGE2) is a potent vasodilator and involved in the renal autoregulation of sodium and water.1Bryson T.D. Harding P. Prostaglandin E2 EP receptors in cardiovascular disease: An update.Biochem Pharmacol. 2022; 195114858https://doi.org/10.1016/j.bcp.2021.114858Crossref PubMed Scopus (12) Google Scholar Through cyclooxygenase-2 (COX-2) inhibition, NSAIDs block the synthesis of prostaglandins (Figure 1 A).2Burnier M. Revue Médicale Suisse : Médicaments : une cause sous-estimée d’hypertension artérielle.Revue Médicale Suisse. 2014; 10: 1661-1665PubMed Google Scholar The reduced level of PGE2 results in systemic vasoconstriction and reduction of renal blood flow followed by sodium and water retention (Figure 1 A). In our case, the patient did not exhibit acute renal damage, likely because of a preserved cardiac output maintaining the renal autoregulation. The systemic arterial hypertension was therefore probably caused by the systemic vasoconstriction.2)PAH is characterized by abnormalities in the pulmonary endothelium resulting in disturbed syntheses of vasodilating and -constrictive substances (nitric oxide (NO), thromboxane A2, endothelin and prostacyclin).3Lang I.M. Gaine S.P. Recent advances in targeting the prostacyclin pathway in pulmonary arterial hypertension.Eur Respir Rev. 2015; 24: 630-641https://doi.org/10.1183/16000617.0067-2015Crossref PubMed Scopus (67) Google Scholar Prostacyclins (PGI2) are primarily synthesized by vascular endothelial cells and a potent vasodilator of the pulmonary arteries.3Lang I.M. Gaine S.P. Recent advances in targeting the prostacyclin pathway in pulmonary arterial hypertension.Eur Respir Rev. 2015; 24: 630-641https://doi.org/10.1183/16000617.0067-2015Crossref PubMed Scopus (67) Google Scholar,4Fredenburgh L.E. Ma J. Perrella M.A. Cyclooxygenase-2 inhibition and hypoxia-induced pulmonary hypertension: effects on pulmonary vascular remodeling and contractility.Trends Cardiovasc Med. 2009; 19: 31-37https://doi.org/10.1016/j.tcm.2009.04.003Crossref PubMed Scopus (28) Google Scholar NSAIDs reduce, by COX-2 inhibition, PGI2 levels in the pulmonary smooth muscle cells inducing pulmonary vasoconstriction and consequently an increase of PVR (Figure 1 B). One can speculate that the increased PVR led in our patient to an increase of the right-to-left shunt. However and unfortunately, TTE did not allow to evaluate changes of the degree of shunting due to a reduced image quality and we refrained from insertion of a Swan-Ganz catheter due to the inherent risk of such catheters in patients with right-to-left shunt.3)Our patient had chronic thrombocytopenia which is typically found in Eisenmenger’s syndrome.5Danioth S. Schanz U. Greutmann M. Thrombocytopaenia in cyanotic CHD.Cardiol Young. 2021; 31: 429-434https://doi.org/10.1017/S1047951120004230Crossref PubMed Scopus (0) Google Scholar Common etiologies for the decline in platelet count were ruled out. We excluded heparin-induced thrombocytopenia, thrombosis, disseminated intravascular coagulation, infectious disease (CMV, EBV, HIV, viral hepatitis, Helicobacter pylori), hypersplenism, folic acid and B12 deficiency, autoimmune etiology. Drug-induced aggravated thrombocytopenia was therefore suspected. Initially, sertraline and macitentan were incriminated but finally excluded as cause as platelet count returned to its usual level despite their continuation throughout the hospitalization. Therefore, ibuprofen remains the only suspected drug. Thrombocytopenia is a known adverse effect of ibuprofen. A case report describes a patient who received a total of 2.6g of ibuprofen over 6 days who developed the 6th day thrombocytopenia at 6 G/L which recovered after discontinuation of the treatment and administration of prednisolone and immunoglobulins (Ig).6Jain S. Ibuprofen-induced thrombocytopenia.Br J Clin Pract. 1994; 48: 51Crossref PubMed Google Scholar A further report describes a 71-year-old patient who developed thrombocytopenia after the administration of 2.4g of ibuprofen who also recovered after prednisolone and Ig administration.7Meyer T. Herrmann C. Wiegand V. Mathias B. Kiefel V. Mueller-Eckhardt C. Immune thrombocytopenia associated with hemorrhagic diathesis due to ibuprofen administration.Clin Investig. 1993; 71: 413-415https://doi.org/10.1007/BF00186632Crossref PubMed Scopus (22) Google Scholar In both cases, an immune-mediated reaction was suspected with antibody-mediated platelet destruction. Consistent with these reports, we observed an improvement of the platelet count in our patient seven days after stopping ibuprofen, however, without the need of corticoid or Ig administration. We can derive three important messages from this case.1)Like in the general population, NSAIDs are frequently prescribed as first-line medication for acute pain in PAH patients.2)Administration of NSAIDs can have potentially deleterious consequences in these patients, particularly on PVR, platelet count and systemic arterial pressure.3)Consequently, prescription of NSAIDs should be avoided in PAH patients similar to other patient populations with cardiovascular disease.