The differentiation of SIDS from accidental or inflicted suffocation may be impossible in some cases. Severe pulmonary intra-alveolar hemorrhage has been suggested as a potential marker for such differentiation. Our aims are to: (1) Compare pulmonary hemorrhage in SIDS and a control group comprised of infants whose deaths were attributed to accidental or inflicted suffocation. (2) Review individual cases with the most severe pulmonary hemorrhage regardless of the cause of death, and (3) Assess the effect of age, bedsharing, cardiopulmonary resuscitation, and postmortem interval, with regard to the severity of pulmonary hemorrhage in SIDS cases. We conducted a retrospective study of all postneonatal cases accessioned by the Office of the Medical Examiner in San Diego County, California who died of SIDS or suffocation between 1999 and 2004. A total of 444 cases of sudden infant death caused by SIDS (405), accidental suffocation (36), and inflicted suffocation (3) from the San Diego SIDS/SUDC Research Project database were compared using a semiquantitative measure of pulmonary intra-alveolar hemorrhage [absent (0) to severe (4)]. Grades 3 or 4 pulmonary hemorrhage occurred in 33% of deaths attributed to suffocation, but in only 11% of the SIDS cases, however, all grades of pulmonary hemorrhage occurred in both groups. Therefore, our results indicate that the severity of pulmonary hemorrhage cannot be used in isolation to determine the cause or manner of sudden infant death. Among SIDS cases, those with a higher pulmonary hemorrhage grade (3 or 4) were more likely to bedshare, and with more than one co-sleeper, than those with a lower pulmonary hemorrhage grade (0 or 1). We conclude that each case must be evaluated on its own merits after thorough review of the medical history, circumstances of death, and postmortem findings.