This study identified the range of anatomical variation of the arterial supply to pyramidalis and reviewed its other morphologic characteristics using GTN-augmented CTA in live female participants for the first time. Additional novelty of this study includes the reporting of pyramidalis prevalence and morphological data in a female cohort of Australian geographic origin. Pyramidalis prevalence data varies across ethnic groups [7, 11, 13, 15, 16, 23, 28], supporting the proposed hypothesis of a genetic predisposition to presence or absence [23]. A systematic review of studies investigating the pyramidalis, hypothesised that the geographic origin in which the study was conducted to be highly relevant [4]. To date, the prevalence of pyramidalis has been investigated in studies from several geographic origins, including India [7, 15, 28], Greece [23], Brazil [13], Africa [16] and Japan [11], but the current study is the first to do so in a study population of Australian geographic origin.

A case of right-sided, unilateral pyramidalis duplication was identified in this study, a finding that has not been reported since 1848 [12]. Previous studies have reported that on occasion duplication can occur with two pyramidalis muscles occurring unilaterally or bilaterally totalling three and four pyramidalis muscles, respectively [15, 33]. Others have observed double pyramidalis muscles on both sides and although very small, all four muscles were distinct [12]. One study reports that the anatomist may rarely meet with three pyramidalis muscles in one subject [33], and others report that there are sometimes two pyramidales on one side, and one on the other [5]. Of 787 participants included in a systematic review and meta-analysis of the anatomical variations of the pyramidalis muscle [4], no cases of duplication of the pyramidalis were recorded.

Our findings for bilateral and unilateral prevalence of the pyramidalis muscle in Australian women of 75% (24/32), and 6.25% (2/32), respectively, is similar to the pooled prevalence estimates recently reported in a systematic review of a multiethnic population of 82.3% for bilateral prevalence, and 4.91% for unilateral prevalence [4]. Bilateral absence in the current study of 18.75% (6/32) is also similar to the pooled prevalence estimates for fresh frozen specimens reported as 11.9% in the said recent systematic review. The pooled prevalence estimates of 5.2% for bilateral absence in a population of human cadaveric specimens that had been formalin fixed, reported in the aforementioned systematic review and meta-analysis on the anatomical variations of the pyramidalis muscle differs from both their fresh frozen population and from our observations. This difference may result from the effect of formalin on cadaveric tissue planes, increasing the risk of tearing the fibres of pyramidalis as they adhere to the posterior aspect of the anterior rectus sheath. The pyramidalis fibres are frequently atrophic with some fatty replacement in the cadaver.

In this study, the most prevalent pattern of arterial supply to pyramidalis was derived from an exclusive, isolated muscular branch of the IEA. To our knowledge, the existence of such a vessel has been proposed by only one other anatomist [6]. In contrast to historically proposed origins of arterial supply to pyramidalis, this study found no examples of pyramidalis arterial origin from the cremasteric (external spermatic artery) branch of the IEA, or its female equivalent (the artery of the round ligament) [6, 20], or the pubic division of the IEA [2]. The present study population is exclusively female, with a mean age of 50 years, since the participant population consisted of female patients undergoing CTA ahead of breast reconstructive surgery using DIEP flap. The arterial supply to pyramidalis could not be reliably discerned by an experienced radiologist from CTA in 22% of cases, despite the 0.5 mm slice thickness and GTN-augmentation. Subjectively, this younger female cohort had no atheroma to interfere with the imaging interpretation, and it is likely the artery diameter was below the resolution of the imaging protocol in these cases. Knowledge of these arterial patterns from this study will aid evaluations of digital subtraction angiography and new technologies such as photon-counting CT angiography, since these imaging modalities may be more accurate than conventional CTA for very small artery identification [9, 30].

We found in our Australian female population, the prevalence of 75% (24/32), unilateral prevalence was 6.2% (2/32) (1 left side, 1 right side) and bilateral absence was 18.8% (6/32) and mean medial border height of 59.6 mm ± 16 mm (95% CI 53 mm to 66.2 mm) on the left and 59.2 mm ± 17.5 mm (95% CI 51.9 mm to 66.4 mm) on the right, to be consistent with recent publications [4]. Morphometric measurements have been reported in geographic regions including India [7, 15, 28], Greece [23], Brazil [13], Africa [16], and Pakistan [1] but none from Australia. Base width has been reported in the literature with great heterogeneity, ranging from 12 mm [15] and 19.9 mm ± 2.3 mm [4, 28], whilst the current study reported a mean base width of 25.2 mm ± 5.1 mm (95% CI 23 mm to 27.3 mm) on the left and 24.4 mm ± 6.1 mm (95% CI 21.9 mm to 27 mm) on the right. The variation may be due to the limitation of CT to precisely resolve the difference between connective tissue structures and adjoining muscle fibres leading to systematic overestimation of muscle base width. In addition, our study was based upon living, younger patients, rather than cadaveric studies, including older subjects. No statistically significant difference was found in medial border height (p = 0.808) or base width p = 0.355). These findings are consistent with prior published studies [23].

Only one other study has investigated pyramidalis thickness [7], however direct comparisons are compromised due to heterogeneity in methods between studies. The prior study [7], conducted analysis on 25 formalin fixed cadavers, including 17 males and 8 females of Indian geographic origin. They measured thickness at the mid-point of the muscle using measuring tape and digital vernier callipers, whilst the current study measured thickness using CTA images of 26 living female participants of Australian geographic origin. In the current study, the measurement was derived from an axial view, at a point 5 mm cephalad to the pubic crest using clinical radiology reporting software.

The pyramidalis is routinely used as a surgical landmark for accurate longitudinal incisions [7, 29]. Furthermore, the muscle is increasingly becoming a tissue of choice for many reconstructive procedures, including occlusion of fistulas [27], free flap procedures [32], colposuspension [31] and recurrent hernia repair [24]. The identification of pyramidalis and its arterial supply using CT in living patients may inform presurgical planning.

The authors acknowledge limitations within this study, specifically a small sample size of n = 32 (n = 50 pyramidalis muscles). In addition, our study population is exclusively female, with a mean age of 50 years ± 7.27 years, since the study is derived from a retrospective study of younger pre-surgical patient cohort. Whilst CT angiography of the lower limbs is also frequently augmented by vasodilators, which would allow inclusion of males, these studies are frequently compromised by atheroma, the indication for the majority of these tests, reducing the sensitivity for small vessel identification, and were not used in this study for this reason. Further studies with a larger sample size, mixed gender, age, and other ethno-geographic groups are recommended. Retrospective studies of digital subtraction angiography and new technologies such as photon-counting CT angiography are likely to improve vessel detection.