Chronic obstructive pulmonary disease (COPD) is, according to the WHO, the fifth leading cause of death worldwide, and is expected to increase to rank third in 2030. Few robust biomarkers for COPD exist, and several attempts have been made to find suitable molecular marker candidates. One rising research area is breath analysis, with several published attempts to find exhaled compounds as diagnostic markers. The field is broad and no review of published COPD breath analysis studies exists yet. We have conducted a systematic review examining the state of art and identified 12 suitable papers, which we investigated in detail to extract a list of potential COPD breath marker molecules. First, we observed that no candidate markers were detected in all 12 studies. Only three were reported in more than one paper, thus reliable exhaled markers are still missing. A major challenge is the heterogeneity in breath sampling technologies, the selection of appropriate control groups, and a lack of sophisticated (and standardized) statistical data analysis methods. No cross-hospital/study comparisons have been published yet. We conclude that future efforts should (also) concentrate on making breath data analysis more comparable through standardization of sampling, data processing, and reporting.