Brachial plexus birth palsy (BPBP) is the most common peripheral nerve injury in children (prevalence <5.1 per 1000 live births) and conservative management is routinely used to manage them. We have previously systematically reviewed studies investigating primary conservative management, published between 1992 and 2002, and found these to be inconclusive. The aim of this review was to update our previous systematic review to include studies published between 2002 and 2008, synthesise the data thus obtained with that of our previous review and create an up-to-date body of evidence for conservative management of infants with BPBP. Fifteen databases were searched systematically for quantitative studies (randomised controlled trials, comparative studies, case series), published in English between January 2002 and June 2008 inclusive. Studies were excluded if they investigated infants who, in their first two years of life, underwent microsurgical repair of the brachial plexus, surgical management of secondary deformities or received other treatments traditionally delivered by surgeons, such as Botulinum toxin injections. The eligibility of each study identified from the database searches was evaluated against the inclusion criteria by two independent reviewers. These studies were then critically appraised for level of evidence using the National Health and Medical Research Council of Australia Hierarchy of Evidence and methodological quality using the Critical Review Form - Quantitative Studies. Data pertaining to the demographic characteristics of study participants, treatments received, main results and outcome measures used were also extracted. Where any disagreement between reviewers occurred, consensus was reached by discussion. Data from the recently published studies were narratively synthesised and then combined with the data gained from our previous systematic review to create a body of evidence on primary conservative management for BPBP infants. Four publications, representing three studies (one comparative study, two case series), were sourced. Methodological quality scores of these studies ranged from 6 to 12 (maximum =16). The current body of evidence (publications from 1992 to 2008) therefore comprises 11 studies, four using a comparative design and seven using a case series design. Six of the 11 studies were classified as being of "poor" methodological quality (score <8). Conservative management mainly consisted of exercise therapy, although splinting, massage and dynamic traction were also used. All studies lacked a clear definition of conservative management sufficient to allow replication of the treatment in a clinical setting. A variety of outcome measures were used, limiting comparability of the studies. Data from the three case studies suggests that conservative management may be more effective in infants with upper and middle plexus injuries compared with total plexus palsy. However, the different outcome measures used and lack of comparison groups limit the strength of this finding. The body of evidence investigating the use of primary conservative management of infants with BPBP remains inconclusive. The studies published to date are limited in number, level of evidence and methodological quality. Further, a variety of outcome instruments, with limited psychometric properties, have been used to investigate management techniques which do not mirror those of contemporary clinical practice.