Prevalence of Sarcopenia in Children Undergoing Total Pancreatectomy with Islet Autotransplantation (TPIAT) and Implications for Short-Term Surgical Outcomes

Abstract

Introduction

Sarcopenia, the loss of muscle mass and function, is a measure of frailty that is correlated with perioperative morbidity in patients with chronic illness and transplant. We sought to understand the prevalence of sarcopenia in our pediatric population undergoing total pancreatectomy with islet cell autotransplantation (TPIAT) and how sarcopenia impacts short-term postsurgical outcomes.

Methods

This was a single-center, IRB-approved study of children (1–18.99 years) who underwent TPIAT at our institution. Patients were identified from a prospective registry where clinical data (pain assessments, 30-day postoperative outcomes) were available near the time of surgery. Sarcopenia was assessed via retrospective analysis of cross-sectional imaging using segmentations of the psoas muscle and total abdominal skeletal muscle. Z-scores (<−1.6) calculated using established normal values defined sarcopenia. Change in muscle mass over time was also determined.

Results

In our cohort, 32.4% (22/68) of patients were sarcopenic at the time of TPIAT by psoas muscle but only 7.4% (6/81) were sarcopenic by total abdominal skeletal muscle. There was no significant difference in primary outcomes (30-day readmissions, reoperations, and complications; and the number of days spent Inpatient after surgery) based on sarcopenia or decreasing muscle mass. There was a statistically significant association between psoas-determined sarcopenia and higher frequency of pain before TPIAT (p=0.016).

Conclusions

Sarcopenia was 32.4% prevalent by psoas muscle measurement but was less prevalent by total abdominal skeletal muscle measurement in children undergoing TPIAT. There was no relationship between the presence of sarcopenia or decreasing muscle mass and 30-day postoperative outcomes but there was a positive association between higher frequency of pre-operative pain and sarcopenia.

Introduction

Pediatric pancreatitis is increasingly recognized ^1,2, with a large study estimating an incidence of 12 in 100,000 children for acute pancreatitis (AP) and 2 in 100,000 children for chronic pancreatitis (CP) ³. In patients with pancreatitis for whom maximal medical and endoscopic therapies fail, surgery may be performed. Total pancreatectomy with islet cell autotransplantation (TPIAT) is a surgical option for patients who are not candidates for, or have failed, conventional surgical procedures. There is a need to better understand risk factors that may impact outcomes in children undergoing TPIAT ^4,5.

Sarcopenia, the state of reduced skeletal muscle mass and function, is a measure of frailty that is of interest in the study of the complications of aging and chronic disease ⁶. Multiple organizations have issued consensus statements recommending diagnostic criteria for sarcopenia based on assessments of skeletal muscle mass and function in adults ^6-11.There is less consensus on the definition of sarcopenia in children, and most pediatric research has relied on skeletal muscle mass as a marker for sarcopenia ¹². Some of the more commonly employed modalities for assessing skeletal muscle mass in children include dual-energy x-ray absorptiometry (DEXA), bioelectrical impedance analysis (BIA), computed tomography (CT), and magnetic resonance imaging (MRI) ^13,14.

Though not consistent, low skeletal muscle mass has been implicated in adverse outcomes in children with conditions including chronic liver disease, cancer, inflammatory bowel disease, and in children who have undergone organ transplant ^12,15-21. In adult patients with pancreatic disease, sarcopenia has been associated with increased frequencies of hospitalization and reduced survival, pancreatic exocrine insufficiency, and poor nutritional status ²²⁻²⁴ One study in adults showed that approximately one third of patients undergoing TPIAT have sarcopenia, sarcopenic patients are more likely to be discharged to residential rehabilitation facilities rather than to home, and importantly, sarcopenia is independently associated with low islet yield ²⁵. There remains a relative lack of literature examining pediatric surgical patients ²⁶ and in particular, pediatric patients with pancreatitis and those undergoing TPIAT.

Our primary aims were to describe the prevalence of sarcopenia in a pediatric sample undergoing TPIAT and to examine associations between sarcopenia and 30-day postoperative outcomes. Our secondary aim was to explore the relationship between sarcopenia and clinical data in the same patient sample. We hypothesized that we would see a high frequency of sarcopenia in our study sample and that sarcopenia would predict outcomes.

Materials and Methods

Demographic and Clinical Data Collection

This was a single-center, IRB-approved study of children who underwent TPIAT at our institution. A total of 131 patients were identified from a prospective registry where data was collected from April 2017 to July 2024. Postoperative outcomes prospectively collected within 30 days of surgery available in the registry included: readmission within 30 days (yes/no), re-operation within 30 days (yes/no), any surgical complication (yes/no), and number of days spent inpatient after surgery. Factors representative of the patients' history of disease were recorded, including diagnosis (acute recurrent pancreatitis (ARP) or CP), use of enteral feeds (gastric or jejunal), use of total parenteral nutrition (TPN), presence of impaired glucose tolerance, serum albumin, pancreatic exocrine insufficiency (PEI), treatment by a pain clinic or pain specialist, and number of times hospitalized in the 18 months prior to surgery. Other recorded data included age and sex.

Pain Data Collection

Prospectively collected survey data pertinent to assessment of pain and quality of life was also available in the registry. Assessments included Pain Assessment with Visual Analog Pain Scale (VAS), and Pain Catastrophizing Scale (Appendix), which are validated tools in this population ²⁷. Both yes/no responses and numeric scores were recorded with the Pain Assessment with VAS, which assessed frequency and severity of pain (questions 1, 3, 6, and 7). Numeric scores were assigned with the Pain Catastrophizing Scale (range 0–52), and patients were also classified as low (0–14), moderate (15–25), or high (26–52) pain catastrophizers according to Pielech et al ²⁸. Partial responses to the Pain Catastrophizing Scale questionnaire were excluded from analysis.

Sarcopenia Assessment

Sarcopenia was determined using clinically-obtained cross-sectional imaging (CT or MRI) available in the medical record prior to TPIAT. Previous research in a pediatric population has demonstrated equivalence of these modalities in assessing sarcopenia by our method ²⁹. The picture archiving and communication system (PACS) at our institution was queried to identify the closest CT or MRI examination prior to and within 3 months of TPIAT. Patients were excluded if no imaging was available within this period of time (Supplemental Figure I, Supplemental Digital Content 1, http://links.lww.com/MPA/B498). The L3 level was identified by a pediatric radiology fellow (MM) and research fellow (PD) for each study and then confirmed by a senior pediatric radiologist (AT; 12 years of clinical experience). An automatic segmentation tool ³⁰ was then applied to generate preliminary segmentations of skeletal muscle at the L3 vertebral level. Preliminary segmentation results were manually adjusted using 3D-Slicer ³¹⁻³³ by either MM or PD to produce two segmentations: psoas muscle only (termed: psoas muscle cross-sectional area [PMA]), and all abdominal skeletal muscle including psoas muscle (termed: skeletal muscle cross-sectional area [SMA]) (Figure I). Adjusted segmentations were then reviewed and further adjusted by the fellow who had not performed the initial adjustment, providing two reader verification.

Figure I.

Axial T2 single shot magnetic resonance image of a 7-year-old male with example segmentations. The psoas muscle cross-sectional area (PMA) segmentation is in red and the total abdominal skeletal muscle cross-sectional area (SMA) segmentation is in green, which includes the red PMA segmentation. This patient met sarcopenia criteria by the PMA measurement (Z-score −2.38) but not the SMA measurement (Z-score 0.50).

PMA and SMA were recorded from the adjusted segmentations. Z-scores for each were calculated using previously established normal values for PMA ³⁴ and SMA ³⁵. Based on the patient age ranges used to establish normal values for PMA and SMA in these prior studies, PMA Z-scores were calculated only for patients aged 1–15.99 and SMA Z-scores were calculated only for patients aged 2–18.99 years of age. A Z-score < −1.6 (to capture the bottom 5th percentile) was used to define the presence of sarcopenia for both groups ³⁶.

Change in Muscle Mass Assessment (Muscle Area Trend)

To understand changes in muscle mass over time in children undergoing TPIAT (as opposed to just single time point definitions of sarcopenia), we further queried our PACS to identify serial imaging before TPIAT. CT or MRI examinations at one-year intervals before surgery (9–15 months, 21–27 months, 33–39 months, 45–51 months, 57–63 months) up to the earliest possible imaging examination were identified. PMA and SMA were measured for each identified examination and Z-scores were calculated. Any patient with cross-sectional studies at two or more time points was included in this analysis (Supplemental Figure 2, Supplemental Digital Content 1, http://links.lww.com/MPA/B498), and the SLOPE function in Excel (Microsoft Corporation, Redmond WA) was used to determine the rate of increase or decrease in muscle mass in the years leading up to surgery. The slope of the regression line was used to define an increase in muscle mass (>l) or a decrease in muscle mass (<l).

Statistical Analysis

Descriptive statistics were used to summarize demographic and basic clinical data. Fisher's exact test and the Mann-Whitney U test were used to analyze categorical and continuous variables, respectively. In addition, univariate linear and logistic regression were performed when appropriate using Statistical Analysis System version 9.4 (SAS Institute, Cary, North Carolina) and MedCalc Statistical Software version 22.009 (MedCalc Software Ltd., Ostend, Belgium). A p-value < 0.05 was considered statistically significant for all inference testing.

Results

Study Sample Description and Frequency of Sarcopenia

Pre-TPIAT Psoas Muscle Area

Of the 131 patients in the registry, 18 were excluded from PMA analysis due to age outside the range of 1–15.99 years and 45 patients were removed due to lack of available imaging. This yielded a total of 68 patients, of whom 34/68 (50%) were female. Median age was 11.1 years (IQR 8.3–13.4 years) and median BMI was 19.2 (IQR 16.4–23.9). 45/68 of the available imaging examinations were CT and 23/68 were MRI. Based on PMA, 22/68 (32.4%) of patients were sarcopenic at the time of TPIAT.

Pre-TPIAT Skeletal Muscle Area

Of the 131 patients in the registry, 5 patients were excluded from SMA analysis due to age outside of the range 2–18.99 years and 45 patients were excluded due to lack of available imaging. This yielded a total of 81 patients, of whom 42/81 were female (51.9%). Median age was 11.6 years (IQR 8.8–14.5 years) and median BMI was 19.9 (IQR 17–24). Based on SMA, 6/81 (7.4%) of patients were sarcopenic at the time of TPIAT.

Among the 22 patients identified as sarcopenic by PMA and the 6 patients identified as sarcopenic by SMA, 6 patients were sarcopenic by both methods, including 2 patients who were 16–18.99 years old. Eighteen patients were sarcopenic by only PMA and no patients were sarcopenic by only SMA.

Muscle Area Trend

The number of patients with available scans at each specified time point prior to TPIAT and the frequency of sarcopenia at each time point, is detailed in Table 1 and shown graphically in Figure 2. There were 63 patients who had at least two serial imaging studies to establish a muscle area trend by PMA, 29 (46%) of whom had a trend of decreasing muscle area prior to TPIAT, and the remainder had increasing muscle area. There were 76 patients with at least two serial imaging studies to establish a muscle area trend by SMA, 27 (35.5%) of whom had a trend of decreasing muscle area, 47 (61.8%) of whom had increasing muscle area, and two of whom had no change in muscle area prior to TPIAT.

Table 1:

Total number of imaging examinations at each time point prior to total pancreatectomy with islet autotransplantation (TPIAT) with sarcopenia counts in both psoas muscle area (PMA) and skeletal muscle area (SMA) groups. Pre-TPIAT imaging was the imaging examination most proximal to TPIAT within the 3 months prior to TPIAT.

		Earliest	57–63 months	45–51 months	33–39 months	21–27 months	9–15 months	Pre-TPIAT
PMA	# of available examinations	46	2	8	15	18	37	68
	Frequency of sarcopenia	18 (39.1%)	2 (100%)	6 (40%)	6 (33.3%)	14 (37.8%)	14 (37.8%)	22 (32.4%)
SMA	# of available examinations	45	2	9	20	23	43	81
	Frequency of sarcopenia	4 (8.9%)	0 (0%)	1 (11.1%)	2 (10%)	1 (4.3%)	4 (9.3%)	6 (7.4%)

Figure 2a, 2b:

(a) Psoas muscle z-scores and (b) skeletal muscle z-scores in patients with decreasing muscle mass from the earliest scan date to the time of total pancreatectomy with islet autotransplantation (TPIAT). The dotted line represents the z-score cutoff for sarcopenia of −1.6.

Pain and Quality of Life Assessment

Summary results for Pain Assessment and Pain Catastrophizing Scale are presented in Supplementary Table 1, Supplemental Digital Content 2, http://links.lww.com/MPA/B499.

Primary Outcomes and Association with Sarcopenia

Pre-TPIAT Psoas Muscle Area

In the PMA group, 10/68 (14.7%) patients were readmitted within 30 days of TPIAT, 7/68 (10.3%) patients had reoperations within 3 months of TPIAT, and 17/68 (25%) patients had a complication within 30 days of surgery (Table 2). There was no statistically significant difference in the frequency of primary outcomes between patients with and without sarcopenia defined by pre-TPIAT PMA. There was also no statistically significant difference in the number of days spent inpatient after surgery in patients with and without sarcopenia by PMA.

Table 2:

Primary outcomes with respect to sarcopenia defined based on the immediate pre-total pancreatectomy with islet autotransplantation (TPIAT) imaging and with respect to muscle mass trend based on imaging in years preceding TPIAT. Binary outcome data are presented as counts and percentages with outcome frequencies calculated as a percentage of the total number of patients in each group and sarcopenia frequencies calculated as a percentage of participants with the outcome of interest. P-values reflect 2×2 comparisons of outcome frequencies among patients without and with sarcopenia using Fisher’s exact test. The continuous outcome of Days Spent Inpatient is presented as medians with interquartile ranges with p-values reflecting comparisons between groups of patients without and with sarcopenia using the Mann Whitney U Test.

		30-Day Readmissions			30-Day Reoperations			30-Day Complications			Days Spent Inpatient
	Total Patients	Total Readmissions	With Sarcopenia	P-value	Tota1 Reoperations	With Sarcopenia	P-value	Total Complications	With Sarcopenia	P-value	No Sarcopenia	Sarcopenia	P-value
Pre-TPIAT PMA	68	10 (14.7%)	4 (40%)	0.717	7 (10.3%)	2 (28.6%)	1.000	17 (25%)	4 (23.5%)	0.551	17 (15-21)	16 (14-25.75)	0.640
Pre-TPIAT SMA	81	10 (12.3%)	0 (0%)	1.000	8 (9.9%)	0 (0%)	1.000	19 (23.5%)	0 (0%)	0.327	18 (15-24)	15.5 (14.25-17.5)	0.262
	Total Patients	Total Readmissions	With Decreasing Muscle	P-value	Total Reoperations	With Decreasing Muscle	P-value	Total Complications	With Decreasing Muscle	P-value	Decreasing Muscle	Increasing Muscle	P-value
PMA	63	9 (14.3%)	5 (55.6%)	0.721	6 (9.5%)	2 (33.3%)	0.678	16 (25.4%)	8 (50%)	0.777	20 (16-24)	16 (15-24.5)	0.188
SMA	76	9 (11.8%)	2 (22.2%)	0.472	7 (9.2%)	4 (57.1%)	0.250	17 (22.4%)	3 (17.6%)	0.088	18 (15-21.5)	18 (15-25.5)	0.585

Pre-TPIAT Skeletal Muscle Area

In the SMA group, 10/81 (12.5%) patients were readmitted within 30 days of TPIAT, 8/81 (9.9%) patients had reoperations within 3 months of TPIAT, and 19/81 (23.5%) patients had a complication within 30 days of surgery (Table 2). There was no statistically significant difference in the frequency of primary outcomes between patients with and without sarcopenia defined by pre-TPIAT SMA. There was also no statistically significant difference in the number of days spent inpatient after surgery in patients With and without sarcopenia by SMA.

Muscle Area Trend

In the PMA trend group, 9/63 (14.2%) patients were readmitted within 30 days of TPIAT, 6/63 (9.5%) patients had reoperations within 3 months of TPIAT, and 16/63 (25.4%) patients had a complication within 30 days of surgery (Table 2). There was no statistically significant difference in the frequency of primary outcomes between patients with and without decreasing PMA. There was also no statistically significant difference in the number of days spent inpatient after surgery in patients based on change in PMA.

In the SMA trend group, 9/76 (11.8%) patients were readmitted within 30 days of TPIAT, 7/76 (9.2%) patients had reoperations within 3 months of TPIAT, and 17/76 (22.4%) patients had a complication within 30 days of surgery (Table 2). There was no statistically significant difference in the frequency of primary outcomes between patients with and without decreasing SMA. There was also no statistically significant difference in the number of days spent inpatient after surgery in patients based on change in SMA.

Univariable Associations Between Sarcopenia, Muscle Area Trend, and Clinical Characteristics and Other Outcomes

Pre-TPIAT Psoas Muscle Area

Presence of sarcopenia based on pre-TPIAT PMA was statistically significantly associated with use of TPN (p=0.038). There was no significant univariable association between pre-TPIAT PMA-defined sarcopenia, other clinical characteristics, or outcomes by logistic or linear regression. As an independent predictor, decreasing PMA trend was not significantly associated with any outcome or characteristic (Table 3).

Table 3:

Univariable associations between sarcopenia determined by psoas muscle area (PMA) on immediate pre-total pancreatectomy with islet autotransplantation imaging or by PMA trend and clinical characteristics and outcomes. Significant p-values are bold.

	PMA Sarcopenia			PMA Decreasing Muscle Area
	Outcome (n=# with sarcopenia)	OR (95% CI)	P-value	Outcome (n=# decreasing)	OR (95% CI)	P-value
Logistic Regression	Readmission 30 days	1.48 (0.37, 5.90)	0.577	Readmission 30 days (n=5)	0.64 (0.16, 2.65)	0.538
	Reoperation 30 days (n=2)	0.82 (0.15, 4.60)	0.822	Reoperation 30 days (n=2)	1.80 (0.31, 10.6)	0.516
	Any complication 30 days (n=4)	0.56 (0.16, 1.99)	0.373	Any complication 30 days (n=8)	0.81 (0.26, 2.52)	0.713
	Enteral feeds - gastric or jejunal (n=12)	1.93 (0.67, 5.52)	0.222	Enteral feeds - gastric or jejunal (n=7)	0.75 (0.24, 2.30)	0.615
	TPN (n=11)	3.20 (1.07, 9.58)	0.038	TPN (n=8)	1.20 (0.38, 3.84)	0.754
	Blood glucose abn (n=2)	1.00 (0.17, 5.93)	1	Blood glucose abn (n=4)	0.37 (0.06, 2.21)	0.272
	PEI (n=8)	0.89 (0.31, 2.54)	0.826	PEI (n=8)	1.63 (0.53, 4.98)	0.396
	Treatment by pain specialist (n=9)	0.46 (0.15, 1.34)	0.154	Treatment by pain specialist (n=16)	0.48 (0.14, 1.70)	0.257
	Pain Assessment Q1 (n=15)	0.46 (0.11, 1.81)	0.263	Pain Assessment Q1 (n=21)	1.65 (0.33, 8.21)	0.54
	Pain Assessment Q3 (n=11)	2.92 (0.77, 11.1)	0.115	Pain Assessment Q3 (n=13)	0.72 (0.22, 2.32)	0.579
	Outcome	Beta	P-value	Outcome	Beta	P-value
Linear Regression	Days Inpatient	−0.56	0.803	Days Inpatient	−1.71	0.43
	# days hospitalized last 18 months	0.63	0.47	# days hospitalized last 18 months	0.61	0.533
	VAS Q6 - Acute attack pain level	−1.45	0.094	VAS Q6 - Acute attack pain level	1.2	0.126
	VAS Q7 - Average pain level	−0.76	0.267	VAS Q7 - Average pain level	0.003	0.997
	PCS Total Score	−0.96	0.624	PCS Total Score	0.86	0.646

There was a statistically significant association between pre-TPIAT PMA-determined sarcopenia and frequency of pain as assessed by Question 2 on the Pain Assessment survey (p=0.016) but no other significant associations with pain measures (Table 4).

Table 4:

Contingency tables for Fisher’s exact test between sarcopenia defined by psoas muscle area (PMA) and skeletal muscle area (SMA) or PMA/SMA muscle trend and responses to Question 2 on the Pain Assessment. Significant p-values are bold.

	PMA Sarcopenia			SMA Sarcopenia			PMA Muscle Mass Trend			SMA Muscle Mass Trend
	With Sarcopenia (n=15)	Without Sarcopenia (n=38)	p-value	With Sarcopenia (n=5)	Without Sarcopenia (n=59)	P-value	With Decreasing Muscle (n=20)	With Increasing Musele (n=29)	P-value	With Decreasing Muscle (n=19)	With Increasing Muscle (n=38)	P-value
"On a daily basis"	7 (47%)	18 (48%)	0.016	3 (60%)	29 (49%)	0.392	9 (45%)	16 (55%)	0.829	11 (58%)	19 (50%)	0.156
"Several days per week"	4 (27%)	0 (0%)		1 (20%)	4 (7%)		3 (15%)	2 (7%)		3 (16%)	3 (8%)
"Several days per month"	2 (13%)	10 (26%)		0 (0%)	14 (24%)		4 (20%)	6 (21%)		1 (5%)	11 (29%)
"Every few months"	2 (13%)	10 (26%)		1 (20%)	12 (20%)		4 (20%)	5 (17%)		4 (21%)	5 (12%)
"A few times per year"	0 (0%)	0 (0%)		0 (0%)	0 (0%)		0 (0%)	0 (0%)		0 (0%)	0 (0%)

Pre-TPIAT Skeletal Muscle Area

The frequency of sarcopenia based on pre-TPIAT SMA was too low to allow for regression analysis. However, there was a statistically significant relationship between presence of sarcopenia and use of jejunal feeds in this group (p=0.0183) (Table 5). No other significant associations between pre-TPIAT SMA-defined sarcopenia or SMA trend and clinical characteristics or outcomes, including pain scores, were noted.

Table 5:

Associations between sarcopenia defined by skeletal muscle area on immediate pre-total pancreatectomy with islet autotransplantation imaging or by SMA trend and clinical characteristics and outcomes. Significant p-values are bold.

		SMA Sarcopenia				SMA Decreasing Muscle Area
	Outcome	Total	With Sarcopenia	Without Sarcopenia	P-value	Total	With Decreasing Muscle	With Increasing Muscle	P-value
Fisher's Exact Test	Gastric feeds	21	3	18	0.338	14	4	10	0.754
	Jejunal feeds	27	5	22	0.018	19	7	12	0.767
	TPN	23	2	21	1.000	19	6	13	1.000
	Blood glucose abnormality	9	1	8	0.533	8	3	5	0.704
	PEI	28	1	27	0.659	22	8	14	0.785
	Treatment by pain specialist	43	4	39	1.000	37	12	25	1.000
	Pain Assessment Q1	60	5	55	1.000	57	19	38	0.240
	Pain Assessment Q3	37	5	32	0.146	35	13	22	0.567
	Outcome	Total	With Sarcopenia	Without Sarcopenia	P-value	Total	Decreasing	Increasing	P-value
Mann Whitney U Test	# days hospitaliz ed last 18 months	18 (15, 24)	15.5 (14, 18)	18 (15, 24)	0.269	18 (15, 24)	18 (15, 22)	18 (15, 26)	0.590
	VAS Q6 - Acute attack pain level	8 (5, 9)	5.5 (0, 10)	8 (5, 9)	0.629	8 (6, 9)	8 (4, 10)	8 (6, 9)	0.719
	VAS Q7 - Average pain level	4 (2,6)	4.5 (4, 7)	4 (2,6)	0.606	4 (3, 6)	5 (2, 7)	4 (3, 6)	0.961
	PCS Total Score	16 (13, 20)	19 (16, 22)	16 (12, 20)	0.261	15 (10, 18)	14 (10.5, 18)	16 (10, 18)	0.403

Discussion

Total pancreatectomy with islet autotransplantation is a procedure that is reserved for children with intractable ARP or CP. Due to the potentially debilitating nature of ARP and CP, we expected to see a high prevalence of sarcopenia in our study sample. In this cohort, the prevalence of sarcopenia varied according to the muscle group measured. There was a 32.4% prevalence of sarcopenia based on pre-TPIAT psoas muscle area (PMA) which is similar to the frequency of sarcopenia reported by Trikudanathan et al, who observed a 1/3 prevalence of preoperative sarcopenia determined by skeletal muscle index in adults undergoing TPIAT ²⁵. However, when defined by total skeletal muscle area (SMA) which includes the psoas in addition to all other abdominal skeletal muscles, the prevalence of sarcopenia was only 7.4% in our sample.

While we observed up to a 32.4% frequency of sarcopenia in our sample, there was no significant difference in primary outcomes (30-day readmissions, reoperations, and complications; and the number of days spent inpatient after surgery) based on the presence of sarcopenia or decreasing muscle mass trend. These findings are consistent with those of Trikudanathan et al, who found no relationship between sarcopenia and incidence of major surgical complications, length of hospital stay, or 30-day readmissions after TPIAT in adults²⁵.While several studies in pediatric patients undergoing transplant have identified poor outcomes in those with sarcopenia ^15,37,38, not all studies have found an association. Jeong et al found no association with preoperative low skeletal muscle mass and poor early postoperative outcomes in pediatric patients undergoing liver transplant ³⁹. Raghu et al found a high prevalence (64%) of sarcopenia in pediatric patients undergoing intestinal transplant, without an association with decreased graft survival or overall survival at one and five years ³⁶. The lack of association between sarcopenia and outcomes in our sample could reflect something unique about either the study sample or TPIAT (versus other surgeries). Alternatively, it is possible that the carefully protocolized postoperative course at our institution may have masked subtle variations in length of hospitalization and readmissions.

While there were not significant associations between sarcopenia and primary outcomes, the presence of sarcopenia based on pre-TPIAT PMA was statistically significantly associated with use of TPN (p=0.038), and the presence of sarcopenia determined by pre-TPIAT SMA was significantly associated with the use of jejunal feeds (p=0.0183). These associations are expected given the significant overlap between malnutrition and sarcopenia ¹³.

Our results also show a statistically significant association between the presence of PMA-determined sarcopenia and higher frequency of pain before TPIAT. Other pain measures, including assessment of pain severity and psychological response to pain (catastrophizing), did not show an association with sarcopenia in our study. It is possible that the observed association reflects false discovery given the large number of tested associations and the lack of association with other pain measures. However, sarcopenia has been associated with worse outcomes in adults with back pain and other chronic pain conditions ^40-43. Severity of pain and multisite pain has also been shown to be potentiated by sarcopenia ⁴³. Similarly, central sensitization and pain catastrophizing can lead to persistent pain post-TPIAT and other surgeries ^4,44. To our knowledge, there are no pediatric studies linking sarcopenia and pain. However, lower physical activity and impaired physical performance have been linked to sarcopenia in children ⁴⁵. Older adults with pancreatitis have a higher risk of persistent pain after TPIAT and persistent pain can lead to declines in muscle mass secondary to immobility which may create a pain cycle ²⁵. Future studies are needed to further understand contributors to pain and examine the interplay between preoperative pain, sarcopenia, and persistent postoperative pain in children with pancreatitis requiring TPIAT.

In all 6 cases of sarcopenia determined by pre-TPIAT SMA in our sample, the PMA measurement was also indicative of sarcopenia. However, there were 18 patients who were sarcopenic by pre-TPIAT PMA alone. We cannot explain this discrepancy, or define truth based on the current study. It is possible that this discrepancy is due to differences in the reference standard populations used to calculate z-scores for PMA and SMA. The population used to calculate the PMA z-score consisted of 779 children aged 1–16 years presenting with trauma to an emergency department in Toronto, Canada; patients with documented chronic medical illness or acute spinal trauma were excluded ³⁴. The population used to calculate the SMA z-score was derived from a larger sample of 8817 children aged 2–18.99 years who underwent abdominal CT in Cincinnati, Ohio; and more stringent exclusion criteria were applied ³⁵. It could also be that there are structural or metabolic differences inherent in the psoas muscle versus other skeletal muscle which cause disproportionate psoas muscle loss in pancreatitis. Using a single cross-sectional muscle area to estimate total body muscle relies on the assumption that the single level and muscle group(s) chosen are representative of the whole body. Prior research has shown measurements at the L3 level to correlate well with whole body muscle in adults ⁴⁶. However, theoretical age-related differences in muscle bulk and distribution in children could be a source of error and underscores the need for further pediatric-specific study in this area. Use of, or correlation with, a different measurement tool that analyzes whole-body composition, such as DEXA, might shed further light on this. Of note, one prior study of children with steatotic liver disease showed moderate to strong correlations between PMA and BIA-derived muscle mass ⁴⁷. In addition to different tools for measuring sarcopenia, there are various Z-score cutoffs that have been used in prior research for estimating sarcopenia, and sarcopenia still lacks a firm consensus definition by quantitative methods. Many studies have used a Z-score < −2.0 as a lower threshold, defining the bottom approximately 2.3% of patients as having sarcopenia ^26,29,38. Other research has expanded the definition to include the bottom 10% of patients ⁴⁸. We chose to use a Z-score < −1.6 to capture the bottom 5% of patients, in line with previous authors ³⁶. Despite this slightly broader, but still accepted, definition of sarcopenia, we found no associations with our primary outcomes which we believe strengthens this portion of our analysis.

Our study has several limitations. First, while our cohort is large for a study of children undergoing an uncommon surgery, our conclusions are limited by a relatively small sample size. Second, while the majority of the data collection was prospective, the imaging examinations and therefore sarcopenia data were retrospectively collected based on clinically performed imaging. Not all patients had imaging at all time points of interest and the performance and timing of imaging was not standardized, impacting our sample size. Third, there are technically two components to defining sarcopenia: assessment of muscle bulk and muscle function. Our study, like many others, defines sarcopenia using muscle bulk data alone. Our results cannot characterize deficits in muscle function. Fourth, our assessment of muscle bulk relies on generalizations about whole body muscle volume derived from muscle cross-sectional areas at a single vertebral level. While there are multiple studies showing validity to this approach, particularly in adults, this has been less well proven in children. Fifth, we explored only 30-day outcomes. Longer-term outcomes (beyond 30 days after surgery) may have yielded different results. Finally, longitudinal preoperative pain data, which was not available for this study, may have provided more information about pain trends in chronic pancreatitis.

In conclusion, sarcopenia was prevalent (32.4%) by psoas muscle measurement in our cohort of pediatric patients undergoing TPIAT, but was less prevalent by total skeletal muscle measurement. The discrepancy in sarcopenia assessment underscores the need for further study and standardization of pediatric sarcopenia definitions. While potentially sample size limited, there was no relationship between the presence of sarcopenia or decreasing muscle bulk and 30-day postoperative outcomes in our sample of children undergoing TPIAT but there was a positive association between higher frequency of pre-operative pain and sarcopenia.