Test–retest reliability of splenic volume assessment by ultrasonography

Abstract

While MRI and CT are the gold standards for assessments of splenic size in clinical settings, ultrasonography is particularly suited due to its portability, cost efficiency and easy utilization. However, ultrasonography is associated with subjective assessment, potentially resulting in increased variation. We used a test–retest design aiming to determine the reliability of splenic measurements assessed by ultrasonography during apnea. In addition, we compared reliability between different equations for volume calculations: Koga, Prolate ellipsoid and Pilström. Twelve healthy participants (6 women) performed two tests separated by 15 min, comprising a maximal voluntary apnea in a seated position. Splenic dimensions were measured via ultrasonography for 5 min before and immediately following apnea. Resting splenic volume displayed high test–retest reliability between tests (Pilström: 157 ± 39 mL vs 156 ± 34 mL, p = .651, ICC = .970, p < .001, CV = 2.98 ± 0.1%; Prolate ellipsoid: 154 ± 37 mL vs 144 ± 43 mL, p = .122, ICC = .942, p < .001, CV = 5.47 ± 0.3%; Koga: 142 ± 37 mL vs 140 ± 59 mL, p = .845, ICC = .859, p < .001, CV = 9.72 ± 1.4%). Apnea-induced volumes displayed similar reliability (127 ± 29 mL vs 129 ± 28 mL, p = .359, ICC = .967, p < .001, CV = 3.14 ± 3.1%). Reliability was also high between equations (Pilström vs Prolate ellipsoid: ICC = .818, p < .001, CV = 7.33 ± 0.3%, bias = − 3.1 mL, LoA = − 46.9 to 40.7 mL; Pilström vs Koga: ICC = .618, p < .01, CV = 11.83 ± 1.1%, bias = − 14.8 mL, LoA = − 76.9 to 47.3 mL). We conclude that splenic ultrasonographic measurements have practical applications during laboratory and field-based research as a reliable method detecting splenic volume change consistently between repeated tests. The Pilström equation displayed similar reliability compared to the prolate ellipsoid formula and slightly higher compared to the Koga formula and may be particularly useful to account for individual differences in splenic dimensions.

Splenic contraction and cardiovascular responses are augmented during apnea compared to rebreathing in humans

“…It is possible that a series of maximal-duration apneas elicit a more durable contraction of the spleen than submaximal apneas. It was recently reported that, following contraction induced by a single maximum-duration apnea in subjects inexperienced in apneic diving, spleen volume had reached the baseline, pre-apnea volume after 5 minutes ( Holmström et al, 2022 ). Also, Schagatay et al (2005) reported that Hb had returned to control after 10 minutes of recovery.…”

“…Spleen volume was estimated from ultrasonic measurements using an equation (Pilström) that has not been directly validated against other measurement methods, such as magnetic resonance imaging or direct measurements of resected spleens. However, the Pilström equation has been validated against other equations used for ultrasonic determinations of spleen size, which in turn have been validated using spleen volume measurements conducted via autopsy ( Lodin-Sundström, 2015 ; Holmström et al, 2022 ). Also, the Pilström equation has been reported to display high test–retest reliability ( Holmström et al, 2022 ), especially when using mean values from repeated measurements (several time points), as in the present study.…”

“…However, the Pilström equation has been validated against other equations used for ultrasonic determinations of spleen size, which in turn have been validated using spleen volume measurements conducted via autopsy ( Lodin-Sundström, 2015 ; Holmström et al, 2022 ). Also, the Pilström equation has been reported to display high test–retest reliability ( Holmström et al, 2022 ), especially when using mean values from repeated measurements (several time points), as in the present study. Furthermore, the Pilström equation has been used to estimate spleen volume in several other studies investigating splenic contraction during apnea ( Schagatay et al, 2005 ; Lodin-Sundström and Schagatay, 2010 ; Richardson et al, 2012 ; Lodin-Sundström, 2015 ; Elia et al, 2021 ), and was therefore chosen in order to make comparisons with previous studies possible.…”

Differential splenic responses to hyperoxic breathing at high altitude in Sherpa and lowlanders

“…Measurements of the maximal splenic length ( L ), thickness ( T ), and width ( W ) were used to calculate spleen volume according to the Pilström equation (Schagatay et al., 2005): $$$$\begin{equation*}{V_{{\mathrm{spleen}}}} = L\pi (WT - {T^2})/3\end{equation*}$$$$The equation describes the difference between two ellipsoids and has been used in similar experiments involving spleen volume assessments in association with apnoea and exercise (Engan et al., 2014; Holmström et al., 2021; Schagatay et al., 2012) and is associated with high reliability between repeated tests (Holmström et al., 2022). Baseline (resting) spleen volume was obtained from the 5‐min period before the hyperoxia by averaging the five splenic measurements collected (Holmström et al., 2022). Thereafter, hyperoxic spleen volume was assessed and determined at the point $${S_{{\mathrm{p}}{{\mathrm{O}}_{\mathrm{2}}}}}$$ increased above 98% saturation, which occurred in both groups by the fourth minute.…”

“…First, a challenge with ultrasonography for determining spleen size is the potential measurement error resulting from subjective identification of relevant landmarks and biological variation associated with pulsatile volume changes (Cesta, 2006). Nevertheless, a recent investigation demonstrated that our method is associated with high reliability, reflected by a low coefficient of variation (2.98 ± 0.1%) and high intraclass correlation coefficient (0.970, P < 0.001) between repeated tests (Holmström et al, 2022). Additionally, spleen size measured by ultrasound exhibit strong correlations with actual spleen size measured from cadavers (Loftus et al, 1999) (Elia et al, 2019(Elia et al, , 2021Engan et al, 2014;Holmström et al, 2019;Holmström, Mulder, et al, 2020;Lodin-Sundström & Schagatay, 2010;Richardson et al, 2008;Schagatay et al, 2001Schagatay et al, , 2005.…”