Pulmonary arterial hypertension (PAH) complicates the course of systemic sclerosis (SSc) and is associated with poor prognosis. The elevation of systolic pulmonary arterial pressure (sPAP) during exercise in patients with SSc with normal resting haemodynamics may anticipate the development of PAH. Exercise echocardiography (ExEcho) has been proposed as a useful technique to identify exercise-induced increases in sPAP, but it is unclear how to clinically interpret these findings. In this systematic review, we summarize the available evidence on the role of exercise echocardiography to estimate exercise-induced elevations in pulmonary and left heart filling pressures in patients with systemic sclerosis. We conducted a systematic review of the literature using MEDLINE, Cochrane Library and Web of Knowledge, using the vocabulary terms: ('systemic sclerosis' OR 'scleroderma') AND ('exercise echocardiography') AND ('pulmonary hypertension'). Studies including patients with SSc without a prior diagnosis of PAH, and subjected to exercise echocardiography were included. All searches were limited to English and were augmented by review of bibliographic references from the included studies. The quality of evidence was assessed by the Effective Public Health Practice Project system. We identified 15 studies enrolling 1242 patients, who were mostly middle-aged and female. Several exercise methods were used (cycloergometer, treadmill and Master's two step), with different protocols and positions (supine, semi-supine, upright); definition of a positive test also varied widely. Resting estimated sPAP levels varied from 18 to 35 mm Hg, all in the normal range. The weighted means for estimated sPAP were 22.2 ± 2.9 mmHg at rest and 43.0 ± 4.3 mmHg on exercise; more than half of the studies reported mean exercise sPAP ≥40 mmHg. The assessment of left ventricular diastolic function on peak exercise was reported in a minority of studies; however, when assessed, surrogate variables of left ventricular (LV) diastolic dysfunction were associated with higher sPAP on exercise. We found very high heterogeneity in the methods, the protocols and the estimated sPAP response to exercise. LV diastolic dysfunction was common and was associated with greater elevation of sPAP on exercise.