An initially balanced randomized trial can become less straightforward if follow-up is incomplete. That issue sits alongside the need to distinguish treatment effects from baseline changes. Primary study
Study outline
Two hundred men with idiopathic oligoasthenoteratozoospermia were allocated to CoQ10, L-carnitine, their combination, or placebo. After three months, 174 completed follow-up. Active groups showed favorable baseline-to-follow-up semen and DNA-fragmentation changes, and the authors described the combination as more favorable. Primary report
Our appraisal: follow every randomized participant
The 26 participants without completed follow-up should not disappear from interpretation. Their distribution and reasons for loss can affect the comparison, especially if they relate to response or tolerability. Full review should identify whether the analysis preserved randomized assignment and how missing outcomes were handled.
A significant within-group change in an active arm, coupled with a nonsignificant change in placebo, does not itself prove that the groups differ. The direct between-group estimate is the relevant test. The same principle applies when claiming the combination is superior to either component.
Even a valid combination advantage would not automatically demonstrate synergy. That requires an interaction framework rather than ranking average improvements.
Semen measures and DNA-fragmentation assays are not live-birth outcomes. Their clinical relevance should be discussed without promising conception or implying that a laboratory response resolves every cause of infertility.
The trial contributes a potentially useful component-comparison design, but its interpretation depends on the full analysis. Publication review should inspect participant flow, baseline-adjusted contrasts, multiplicity, and clinical outcome reporting. It should not assign the combined result to L-carnitine alone or assume a different commercial formulation reproduces the study exposure.
Frequently asked questions
How many men were allocated?
Two hundred.
How many completed follow-up?
One hundred and seventy-four.
Were both component-only groups included?
Yes.
Does significance in one arm but not another prove a difference?
No.
Does combination superiority automatically establish synergy?
No.
Are semen measures equivalent to live births?
No.
What should be reviewed first?
Participant flow and direct, prespecified treatment contrasts.