Study selection

The initial search in Medline, CINAHL, and Cochrane yielded 2014 total articles for the categories of trauma, blood pressure, temperature, air travel, and behavior change. After removing duplicates, 1856 titles and abstracts were screened by two independent reviewers. One hundred and forty-eight articles were included for full-text review by two independent reviewers. Disagreements between the two reviewers, whether during abstract or full-text reviews, were addressed through discussion focused on the inclusion and exclusion criteria. When consensus could not be met, a resolution was sought from a third reviewer. Thirty-seven articles were included in this study (PRISMA Fig. 1).

Evidence map of trauma as a risk factor

The results for trauma to the ipsilateral quadrant as a risk factor for the manifestation or progression of BCRL are presented in the bubble map (Fig. 2) and Table 1. The bubble map synthesizes 10 studies relevant to traumas including blood draws, skin puncture, and surgery. Table 1 summarizes 17 studies relevant to traumas including blood draws, skin punctures, surgery, seromas, other trauma events, heavy exercise, and wound infection.

Fig. 2

Precautionary risk of trauma on breast cancer survivors and BCRL

Table 1 Trauma study characteristics and outcomesBlood draws

The three included studies that focused on blood draws as a risk factor for the manifestation of BCRL were prospective cohort studies [8, 11, 17] and included 572 participants at risk for BCRL. The studies did not find a significant association between arm volume change and experiencing one or more blood draws [11, 17]. However, Kilbreath et al. [8] highlighted in their study the insignificant increased odds of manifesting BCRL from a univariate analysis (OR 2.0, 95% CI 0.8–5.2, p = 0.17).

Skin puncture

Six studies were included that looked at skin puncture including injections, fluid aspirations, and acupuncture as risk factors [8, 11, 17, 33,34,35]. Four of the included studies were prospective cohort studies [8, 11, 17, 34], one case–control [35], and one case series [33]. Three hundred and thirty-two of the participants were at risk for BCRL and 51 had BCRL. Three [8, 11, 17] of the four studies that looked at injections did not find a significant association between limb volume and injections in the at-risk limb. Conversely, Clark et al. [34] found 44.4% (N = 8) of those who had a skin puncture developed BCRL (OR 2.44, 95% CI 1.33–4.47). Swenson et al. [35] found that patients with BCRL were more likely to have experienced multiple fluid aspirations after their breast cancer surgery compared to those who did not develop BCRL (OR 1.88, p = 0.005); however, the odds ratio decreased to 1.49 (95% CI 0.73–3.02, p = 0.273) when this risk factor was entered into a multivariate analysis. Acupuncture in the ipsilateral limb did not change or improve limb volume measurements in those at risk for or with BCRL; however, patients reported a subjective improvement in their degree of lymphedema after acupuncture treatment [33].

Surgery

Six studies looked at surgery as a risk factor for the manifestation or progression of lymphedema [35,36,37,38,39,40]. Four studies included surgery to the ipsilateral shoulder, arm, and/or hand [35, 36, 39, 40] and two studies included breast reconstruction [37, 38]. Three of the included studies were retrospective cohort studies [36,37,38], one prospective cohort study [39], one case-controlled, [35], and one case study [40]. Fifty-seven participants had BCRL, 1141 were at risk for BCRL, and 239 were either at risk for or had BCRL. Conflicting evidence exists as to whether surgery to the ipsilateral limb manifests or progresses lymphedema. One study found that 3.8% (4/103) of participants at risk for BCRL developed BCRL after hand surgery [36]. Two studies reported patients with BCRL had a temporary progression of their BCRL but it resolved 2 weeks post-op arthroscopic shoulder repair [40] and a median of 33 months after carpal tunnel release [39]. Conversely, Swenson et al. did not find surgery to the ipsilateral limb to be a risk factor for the manifestation or progression of BCRL (OR 0.94, p = 0.862). Two studies found that breast reconstruction did not have a significant impact on the incidence of BCRL regardless of axillary intervention and reconstruction type [37, 38].

Seroma

Two studies looked at the development of a seroma as a risk factor for the manifestation of progression of lymphedema [41, 42]. One study was a prospective cohort [41] and the other was case-controlled [42]. Ten participants had BCRL and 602 were at risk for BCRL. Conflicting evidence exists as Ribeiro-Pereira et al. found patients at risk for BCRL who developed a seroma demonstrated a higher incidence to lymphedema (HR 1.46, 95% CI 1.14–1.87, p = 0.003), while Shahpar et al. [42] found that the presence of a seroma did not demonstrate a significant correlation with the development of BCRL (OR 0.85, 95% CI 0.4–1.81, p = 0.679).

Trauma events

Eight studies looked at trauma events as potential risk factors for the manifestation or progression of BCRL [8, 11, 17, 35, 42,43,44,45]. The included studies consisted of one randomized controlled trial [44], four prospective cohort studies [8, 11, 17, 45], two case-controlled [35, 42], and one case study [43]. The trauma events ranged from less serious (e.g., bruises) to more serious (e.g., fractures). Twenty-two participants had BCRL, 242 were at risk for BCRL, and 885 were either at risk for or had BCRL. Six of the eight articles did not report a trauma event resulting in the manifestation or progression of BCRL [8, 11, 17, 35, 44, 45]. Conversely, Shahpar et al. [42] found trauma to be a risk factor in the development of BCRL (OR 2.71, 95% CI 1.16–6.33, p = 0.02) and Li et al. [43] found prolonged sun exposure to result in BCRL of the chest wall.

Heavy exercise

One retrospective cohort study looked at heavy exercise (> 80% one repetition maximum) as a risk factor for the manifestation of BCRL (N = 149) [46]. No associations were found between performing heavy resistance training and the development of BCRL.

Wound infection

One prospective cohort study examined the relationship between wound infection and lymphangitis with the development of BCRL [47]. In a sample of 455 participants with BCRL, more than half had reported a prior wound infection (OR 3.11, 95% CI 1.41–6.82, p = 0.003) and lymphangitis (OR 3.83, 95% CI 1.57–9.34, p = 0.002) in the ipsilateral arm.

Evidence map of blood pressure as a risk factor

The results for blood pressure measurements to the ipsilateral limb as a risk factor for the manifestation or progression of BCRL are presented in the bubble map (Fig. 3) and Table 2 and include three studies [8, 11, 17]. All three studies were prospective cohort studies and included a total of 941 participants at risk for BCRL and 5 with BCRL. Having one or more blood pressure measurements to the ipsilateral limb was not associated with an increase in limb volume [8, 11, 17], even in participants who had undergone axillary lymph node dissection (ALND) [17].

Fig. 3

Precautionary risk of taking blood pressure on breast cancer survivors and BCRL

Table 2 Blood pressure study characteristics and outcomesEvidence map of temperature as a risk factor

The results for the impact of temperature on the ipsilateral limb as a risk factor for the manifestation or progression of BCRL are presented in the bubble map (Fig. 4) and Table 3. Six articles relevant to temperature spanning cryotherapy, climate, hot tub use, and sauna use were included. The study types include one randomized controlled trial [48], four prospective cohort studies [8, 45, 49, 50], and one case-controlled [35]. Eighty-four participants had BCRL, 131 were at risk for BCRL, and 280 had BCRL or were at risk for BCRL. Askary et al. [48] found that pulsed local cryotherapy to the ipsilateral limb resulted in a decrease in thickness and circumferential limb difference at the wrist, below the elbow, and above the elbow after 6 weeks (p < 0.01) and 12 weeks (p < 0.001) of treatment compared to the control group. Three studies found that climate did not significantly impact limb volume in terms of manifestation of the progression of BCRL [8, 45, 50]. Conversely, Czerniec et al. [49] found a correlation between weather and arm volume where the maximum temperature on the day prior to limb volume measurements affected extracellular fluid (r = 0.27, p < 0.001), arm volume (r = 0.23, p < 0.001), and self-reported swelling (r = 0.26, p < 0.001). Two studies found hot tub use was not a significant risk factor in the manifestation or progression of BCRL [35,