The Magnetic Resonance Imaging of the Psoas Muscle Area as the Prognostic Factor for Survival and Adverse Events in Spinal Metastasis Surgery

Abstract

Study Design

Retrospective study.

Objective

This study aimed to evaluate the ability of the mortality and adverse events prediction following metastatic spinal surgery of MRI-based cross-sectional psoas muscle area (PMA).

Methods

A retrospective chart review, 120 patients who had undergone metastatic spinal surgery were included. The cross-sectional area identified the PMA under MR-imaging at the L3 or L4 pedicle level, which was classified into 3 tertiles. We used univariate and multivariate cox proportional hazard regression to assess whether PMA was associated with 30-day, 90-day, 1-year, and overall mortality.

Results

The small psoas tertile group populations had a higher mortality rate than the large psoas tertile group. PMA in T1 and T2 had a probability of a higher 90-d mortality rate than PMA in T3 (T1 VS T3: P = .29 and T2 VS T3: P = .12). The median survival time was 7 months, 9 months, and 10 months in PMA T1, T2, and T3, respectively. PMA in tertile 2 had a significantly higher mortality rate of 38% compared to PMA in tertile 3 (HR 1.38, 95% CI .83-2.32, P = .02). Considering PMA as a continuous variable, every 1 mm² increment of PMA resulted in the increase survivorship of 1% (HR .99 with 95% CI .99-1)

Conclusion

The MRI-based cross-sectional PMA tends to predict the 90-d mortality rate and overall mortality rate in spinal metastasis patients who underwent spinal surgery. The PMA should be considered one of the prognostic factors in the treatment of metastatic spinal patients

Keywords

spinal metastasis prognostic factor psoas muscle area mortality MR-image

Introduction

The spine is the most common site of bone metastases, accounting for approximately 50% of cases, and spinal metastasis occurs in about 5% to 10% of patients with primary cancer (Choi SH et al., 2020; Parkin DM et al., 2000; Ware JE Jr, 2000).^1-3 The survival rates of metastatic spinal tumors are increasing with improved diagnostic modality and treatment. However, the mortality and adverse events in metastatic spinal surgery were still high, so the preoperative scoring systems may guide clinical decision-making and stratify therapies regarding surgery or palliative care. In 2015s, Dardic et al studied the prognostic scoring systems for spinal metastases and showed Bauer or modified Bauer has an impact on predicting the remaining survival in patients with spinal metastases. Also, the New England Spinal Metastasis Score (NESMS) and Skeleton Oncology Research Group machine learning algorithms (SORG-MLAs) had a good potential utility of predicting major mortality for decision-making in the clinical guide of treatment following metastatic surgery (Park S et al., 2015; Lee CH et al., 2015; Hibberd CS and Quan GMY, 2017; Balain B et al., 2013; Tokuhashi Y et al., 2014; Wibmer C et al., 2011; Dardic M et al., 2015; Schoenfeld AJ et al., 2016; Karhade AV et al., 2020).^4-12

Sarcopenia is a progressive and generalized skeletal muscle disorder that is associated with an increased likelihood of adverse outcomes, including falls, fractures, physical disability, and mortality. Low muscle strength is the key parameter of sarcopenia (Cruz Jentoft et al., 2019).¹³ The computed tomography-based cross-sectional psoas muscle area (PMA) has been proposed as a useful prognostic factor for predicting mortality and adverse events in spinal metastasis surgery. Dohzono et al showed low CT-based PMA (0–25%) is a poor prognostic factor of survival rate in cervical spinal metastatic surgery (Dohzono et al, 2019).¹⁴ Zakaria et al classified the PMA into 3 tertiles. The tertile-1 (low muscle mass) can predict 90 days and overall mortality in metastatic spinal surgery. Hu MH. et al reported that CT-based PMA is a prognostic factor for 90-day survival and improves the discriminatory ability of the earlier-proposed preoperative scoring system in Asian population (Hu MH et al, 2022).¹⁵

Most of published studies evaluated the CT-based cross-sectional PMA but there was no published study which evaluated the MRI-based cross-sectional PMA for prediction the mortality and adverse events in spinal metastatic surgery. Most of spinal metastatic patient commonly underwent the MRI of spine for preoperative evaluation. This study aimed to evaluate the ability in the mortality and adverse events prediction following spinal metastatic surgery of MRI-based cross-sectional PMA and the correlation between CT-based and MRI-based cross-sectional PMA for prediction the mortality.

Methods

The data were obtained from the medical database of the Department of Orthopedic Surgery, Faculty of Medicine Siriraj Hospital, Mahidol University. We identified consecutive spinal metastasis patients who underwent palliative spinal surgery, between 2009 and 2020. The study was approved by the Siriraj Institutional Review Board (SIRB) (COA No. 569/2563(IRB1)) The informed consent have been deemed exempt from the SIRB.

The inclusion criteria are patient who aged more than 18 years old, patient who has histopathological confirmed spinal metastatic disease, patients who underwent MRI that could be used for morphometric analysis cross-sectional area of the psoas muscle. The exclusion criteria are patients who had more than one primary cancer origin; patients who had any congenital disorder; patients with psychosis or impaired cognitive function; patients who did not previously receive radiotherapy, chemotherapy, or surgery.

Data Source and Variables

We collected demographic data of patients (age, sex, number of spinal metastasis, primary of cancer type, internal organ metastasis, underlying disease), preoperative status (Frankel score, American Society of Anesthesiologists Classification (ASA), The Eastern Cooperative Oncology Group performance status (ECOG)), postoperative status (Level of surgery), and prognostic score (SORG, Nomogram, Modified-Bauer score, NESMS). Morphometric analysis of PMA at the L3 pedicle (or L4 pedicle if the tumor had involved each of the L3 pedicle) was performed by two spinal surgeons at Siriraj Hospital. The PMA measurement was performed twice and 3 weeks apart. The PMA was calculated by cross-sectional area in axial view of magnetic resonance imaging, which selected closely to the date of surgery. Figure 1 illustrates a patient with PMA under cross-sectional magnetic resonance imaging and the location of measurement. The primary outcomes were postoperative 30-d, 90-d, and overall mortality from the date of surgery. Secondary outcomes were adverse events 30-d, 90-d, and overall morbidity, defined as any of the following, surgical site infection (Badhiwala JH et al., 2021)¹⁶ surgical site dehiscence (SSD), pulmonary embolism (PE), deep vein thrombosis (DVT), pneumonia, anemia, upper gastrointestinal bleeding (UGIB), myocardial infarction (MI), anemia and electrolyte imbalance.

Figure 1.

Methodology for psoas measurements.

Quantitative Variables and Statistical Methods

Measurement of the PMA (Figure 1) were divided into 3 equal tertiles (tertile 1 (T1): small muscle area; tertile 2 (T2): medium muscle area; tertile 3 (T3): large muscle area) to provide a clinically meaningful interpretation of the associated hazard ratio (HR). To account for potential sex differences, the 3 tertiles were calculated independently for both genders. The male mean PMA ±standard deviation, with minimum and maximum (range) were 504 mm² ± 109.22, 270–654 mm² for T1; 789.75 mm² ± 79.64, 664–905 mm² for T2; 1078.20 mm² ± 172.98, 908–1634 mm² for T3. For female, measurements were 359.77 mm² ± 45.59, 264–416 mm² for T1; 533.19 mm² ± 59.16, 417–630 mm² for T2; 732.67 mm² ± 74.70, 632–851 mm² for T3. Demographic data compares the 3 psoas tertiles with respect to preoperative characteristics, intra-operative characteristic, preoperative scoring system, and adverse events. Binary outcomes were compared with chi-square tests. When the 3 tertiles statistically significantly differed for any given variable in Table 1, then subsequent calculations would include pairwise comparisons for binary and ordinal variables or Bonferroni corrections for continuous variables. We also analyzed the PMA as a continuous variable. Means of continuous outcomes were first analyzed with Bartlett’s test. If the variance ratios were not significantly different, the outcomes were compared with an analysis of variance. If the variance ratios were statistically significantly different, the nonparametric data were compared with a Kruskal–Wallis test (Ghogawala Z et al., 2016).

Table 1.

Demographic Data.

	Psoas Tertile 1 (N = 40)	Psoas Tertile 2 (N = 40)	Psoas Tertile 3 (N = 40)	P value
Male gender	22 (55%)	22 (55%)	22 (55%)	1.00
Mean age at surgery ± standard deviation	62.63 ± 10.36	56.93 ± 11.74	58.23 ± 13.24	.08
Past medical history/functional status
•ASA score
- 1	4 (44.40%)	1 (11.10%)	4 (44.40%)	.63
- 2	26 (31.70%)	34 (41.50%)	22 (26.80%)
- 3	10 (34.50%)	5 (17.20%)	14 (48.30%)
•Underlying disease
- Hypertension	19 (33.9%)	17 (30.4%)	20 (35.7%)	.79
- Diabetic mellitus	11 (45.80%)	2 (12.50%)	10 (41.70%)	.51
- Lung disease	2 (50%)	1 (25%)	1 (25%)	.77
- Chronic kidney disease	2 (33.30%)	2 (33.30%)	2 (33.30%)	1.00
- Thyroid disease	1 (14.30%)	5 (71.40%)	1 (14.30%)	.09
- Stroke	1 (100%)	0 (0%)	0 (0%)	.37
- Liver disease	1 (33.30%)	0 (0%)	2 (66.70%)	.36
- Heart disease	4 (57.10%)	3 (42.90%)	0 (0%)	.14
- Autoimmune disease	2 (100%)	0 (0%)	0 (0%)	.13
- Other	3 (37.5%)	2 (25%)	3 (37.5%)	.87
Primary cancer
- Lung	11 (44%)	6 (24%)	8 (32%)	.34
- Breast	5 (25%)	7 (35%)	8 (40%)
- Thyroid and ENT	1 (14.30%)	4 (57.10%)	2 (28.60%)
- GI tract	3 (21.40%)	8 (57.10%)	3 (21.40%)
- Renal	0 (0%)	1 (33.30%)	2 (66.70%)
- Male genitourinary tract	11 (45.8%)	6 (25%)	7 (29.20%)
- Female genitourinary tract	4 (50%)	1 (12.50%)	3 (37.50%)
- Hematology	0 (0%)	0 (2%)	2 (100%)
- Others (ewing, adenoid etc.)	0 (0%)	2 (50%)	2 (50%)
- Unknown	5 (38.50%)	5 (38.50%)	3 (23.10%)
Internal organ metastasis	7 (17.1%)	10 (25%)	11 (18.2%)	.48
Preoperative neurological status (Frankel score)				.81
- A	1 (50%)	0 (0%)	1 (50%)
- B	6 (35.30%)	4 (23.50%)	67 (41.20%)
- C	15 (33.30%)	15 (33.30%)	15 (33.30%)
- D	13 (36.10%)	11 (30.60%)	12 (33.30%)
- E	5 (25%)	10 (50%)	5 (25%)
Preoperative scoring system
•SORG				.66
- Good	1 (2.50%)	0 (0%)	0 (0%)
- Intermediate	6 (15%)	4 (10.30%)	5 (12.80%)
- Poor	33 (82.5%)	35 (89.7%)	34 (87.2%)
•Nomogram	219.50 ± 52.07	222.54 ± 52.73	210.56 ± 48.93	.56
•Modified-Bauer score				.30
- 0	3 (7.50%)	0 (0%)	0 (0%)
- 1	12 (30%)	13 (33.30%)	9 (23.10%)
- 2	19 (47.50%)	17 (43.60%)	19 (48.70%)
- 3	6 (15%)	7 (17.90%)	9 (23.10%)
- 4	0 (0%)	2 (5.10%)	2 (5.10%)
•NESMS				.21
- 0	3 (7.50%)	5 (12.80%)	7 (17.90%)
- 1	20 (50%)	12 (30.80%)	8 (20.50%)
- 2	13 (32.50%)	17 (43.60%)	18 (46.20%)
- 3	4 (10%)	5 (12.80%)	6 (15.40%)
Level of metastasis
- Cervical	4 (20%)	10 (50%)	6 (30%)	.44
- Thoracic	17 (24.60%)	25 (36.20%)	27 (39.10%)	.78
- Lumbar	26 (40%)	18 (27.7%)	21 (32.3%)	.02
- Sacral	4 (50%)	3 (37.50%)	1 (12.50%)	.14
Surgical spinal level
- Cervical	2 (18.20%)	5 (45.50%)	4 (36.40%)	.50
- Thoracic	12 (25.50%)	19 (40.40%)	16 (34%)	.59
- Lumbar	19 (38.8%)	14 (28.60%)	16 (32.70%)	.65
- Sacral	5 (41.70%)	3 (25%)	4 (33.30%)	.20
- Illiac	40 (33.30%)	40 (33.30%)	40 (33.30%)	.24
Post-operative medical complications
- Electrolyte imbalanaces	10 (66.70%%)	2 (13.30%)	3 (20%)	.01
- UTI	7 (33.30%)	9 (42.90%)	5 (23.80%)	.50
- Anemia	6 (35.30%)	5 (29.40%)	6 (35.30%)	.93
- Pneumonia	4 (66.70%)	1 (16.70%)	1 (16.70%)	.20
- Pulmonary embolism	0 (0%)	0 (0%)	1 (100%)	.37
- Heart complication	1 (2.4%)	0 (0%)	1 (2.6%)	.60
- Shock	0 (0%)	3 (75%)	1 (25%)	.16
- UGIB	0 (0%)	1 (100%)	0 (0%)	.36
- DVT	1 (50%)	1 (50%)	0 (0%)	.60
- Surgical site infection	1 (20%)	3 (60%)	1 (20%)	.43
- Wound dehiscence	0 (0%)	1 (50%)	1 (50%)	.60
- Other	1 (50%)	1 (50%)	0 (0%)	.61
Surgical complications	2 (33.30%)	3 (50%)	1 (16.70%)	.60

The bold use when P<0.05

The mean survival time at overall time was computed using Kaplan–Meier estimates for all patients in 3 tertiles. The log-rank test was used to assess differences among the 3 Kaplan–Meier curves. We utilized a Cox regression analysis to calculate an estimate of the HRs of overall mortality. The variables with the P-value of less than .2 in univariable analysis were included into the multivariable analysis. We also used the backward elimination in the variable selection. Candidate variables were removed step by step until its P-value below a stopping rule; P-value of .05.

We also assessed the measurement reproducibility by using intraclass correlation coefficient (ICC) and also compare the correlation between measuring PMA under MRI and CT. Reliability was also determine by calculating Cronbach’s alpha. The internal consistency estimates of a >.70 were considered acceptable for group comparisons (Koo TK et al., 2016).¹⁷

All statistical analysis was performed by using StataCorp. Version 17. College Station, TX: StataCorp LLC; 2021 and PASW Statistics for Windows, Version 18.0. Chicago: SPSS Inc; 2009

Results

Participants and Descriptive Data

A total of 655 patients were included in the study, with 535 being excluded due to the lack of an MR image. This left a final sample of 120 patients. Demographic data was collected for these participants, including gender (66 males, 55%), age (mean age of 62.63 + 10.36 for T1, 56.93 + 11.74 for T2, and 58.23 + 13.24 for T3, P = .08), ASA score (ASA 2 was predominant in each tertile, with 26 [31.70%] in T1, 34 [41.50%] in T2, and 22 [26.80%] in T3), underlying diseases (the most common being hypertension, present in 19 [33.9%], 17 [30.4%], and 20 [35.7%] for T1, T2, and T3, respectively, P = .79, followed by diabetic mellitus, present in 11 [45.8%], 2 [12.5%], and 10 [41.7%] for T1, T2, and T3, respectively), primary cancer (the most common being lung cancer, present in 11 [44%], 6 [24%], and 8 [32%] for T1, T2, and T3, respectively, in both genders; male genitourinary tract cancers, including prostate and bladder, were the second most common, present in 11 [45.8%], 6 [25%], and 7 [29.2%] for T1, T2, and T3, respectively), preoperative neurological status (Frankel C was most prevalent in all tertiles, P = .81), and preoperative scoring systems (PSS) including SORG, Nomogram, Modified-Bauer score, and NESMS. The level of metastasis, surgical spinal level, and postoperative medical and surgical complications were also recorded. The demographic details are shown in Table 1.

90-d Mortality

The T1 and T2 groups had a higher tendency for mortality compared to T3, with a P value of .30. When comparing T1 vs T3 and T2 vs T3, the 90-day mortality rate was higher in T1 and T2 compared to T3, with P values of .29 and .12, respectively (Table 2).

Table 2.

90-d Mortality.

	Psoas Tertile 1	Psoas Tertile 2	Psoas Tertile 3	Total	P value
Death	9 (25%)	12 (30%)	6 (15.40%)	27 (23.50%)	.30
Alive	27 (75%)	28 (70%)	33 (84.60%)	88 (76.50%)

Psoas Tertiles	P value
Tertile 1 vs Tertile 2	.62
Tertile 1 vs tTertile 3	.29
Tertile 2 vs tTertile 3	.12

365-d Mortality

There was no difference in 365-day mortality between groups (P = .87) (Table 3).

Table 3.

365-d Mortality.

	Psoas Tertile 1	Psoas Tertile 2	Psoas Tertile 3	Total	P value
Death	13 (48.10%)	12 (42.90%)	13 (41.90%)	38 (100%)	.87
Alive	14 (51.90%)	16 (57.10%)	18 (58.10%)	48 (100%)

Psoas Tertiles	P value
Tertile 1 vs Tertile 2	.69
Tertile 1 vs Tertile 3	.63
Tertile 2 vs Tertile 3	.92

Overall Mortality

The Kaplan–Meier survival analysis of patients in the 3 different tertiles was presented in Figure 2, showing that patients in T3 (Large PMA) had a better survival rate, while patients in T1 had worse survival compared to T2 and T3. The mean survival time was 7 months for T1, 9 months for T2, and 10 months for T3. However, there was no significant difference between tertiles (log-rank test P = .56).

Figure 2.

Kaplan–Meier curve for overall survival in different tertiles. The log-rank test was not significant with a P-value of .56 but the mean survival rate in each tertiles has a tendency of high mortality rate in low psoas tertile. Blue lines indicate T1 (low PMA), Red line indicate T2 (medium PMA), and green lines indicate T3 (large PMA).

Measurement Reproducibility

The Cronbach’s alpha was acceptable for our study, the ICC for correlation between measurement under CT and MRI was .76 (95% CI, .66–.84), the ICC for intra-observer reliability was .986 (95% CI, .98–.99) and the ICC for inter-observer reliability was .95 (95% CI, .908–.973) (Supplementary table)

Univariable and Multivariable Cox Regression

The univariable cox regression analysis and multivariable cox regression were performed to calculate the crude HR and adjusted HR, respectively, for (Choi SH et al., 202)¹ PMA (Parkin DM et al., 1999),² Frankel score (Ware JE Jr, 2000)³ Gender, (Park S et al., 2015)⁴ Age, (Lee CH et al., 2015)⁵ Comorbidity (Hibberd CS and Quan GMY, 2017),⁶ ASA classification (Balain et al., 2013),⁷ Primary cancer (Tokuhashi et al., 2014),⁸ PSS (Modified-Bauer score, SORG, Nomogram, NESMS) which showed on Table 4 and Figure 3.

Table 4.

Univariable and Multivariable Cox Regression.

	Crude Hazard Ratio (95% CI)	P value	Adjusted Hazard Ratio (95% CI)	P value
Psoas muscle area	.99 (.99–1.00)	.48	.99 (.99–1.00)	.21
Psoas muscle area (Tertile 3 (ref))^a
- Tertile 1	1.28 (.79–2.06)	.31	1.21 (.74–2.08)	.49
- Tertile 2	1.07 (.66–1.72)	.78	1.38 (.83–2.32)	.02
Frankel score (E (ref))
- A + B	2.94 (1.42–6.07)	<.01	1.02 (.35–2.99)	.98
- C + D	1.82 (1.03–3.23)	.04	2.13 (1.1–4.09)	<.01
Male	1.40 (.95–2.09)	.09	.84 (.49–1.42)	.51
Age	1.01 (.99–1.03)	.19	.97 (.95–.99)	<.01
Comorbidity (underlying disease)	1.45 (.97–2.18)	.07	1.95 (1.19–3.19)	.01
ASA (1 (ref))
- 2	1.80 (.78–4.17)	.17	.64 (.24–1.69)	.37
- 3	2.93 (1.19–7.27)	.02	1.22 (.42–3.55)	.71
Primary CA (lung (ref))	1.25 (.78–2.01)	.35	.8 (.44–1.46)	.47
Modified Bauer score (4(ref))
- 0	3.96 (.78–20.07)	.01	.25 (.03–2.05)	.19
- 1	2.20 (.67–7.24)	.20	.92 (.22–3.96)	.91
- 2	2.00 (.62–6.47)	.25	.99 (.26–3.81)	.99
- 3	.71 (.20–2.47)	.60	.72 (.19–2.72)	.63
SORG (3 (ref))
- 1	1.81 (.25–13.21)	.55	4.96 (.52–47.33)	.16
- 2	.55 (.30–.99)	.04	1.2 (.59–2.46)	.62
Nomogram	1.01 (1.00–1.01)	<.001	1.00 (.99–1.01)	.07
NESMS (3(ref))
- 0	9.60 (3.96–23.28)	<.001	11.16 (2.27–54.91)	<.01
- 1	6.10 (2.86–12.99)	<.001	7.08 (2.01–24.94)	<.01
- 2	2.47 (1.19–5.11)	<.01	2.64 (.89–7.76)	.08

^aPsoas muscle area which classify as tertile was not include in the multivariable cox regression.

Figure 3.

Forrest plot of adjust hazard ratio.

The PMA T1 and T2, PMA T3 as a reference, showed adjusted HR 1.21 (95% CI .74–2.08; P = .49) and 1.38 (95% CI .83–2.32; P = .02) respectively. The PMA T2 had a significantly high mortality rate when compared to PMA T3 by adjusted HR. Considering PMA as a continuous variable, every 1 mm² increment of PMA resulted in the increase survivorship of 1% (HR .99 with 95% CI .99–1)

Secondary Outcome

The postoperative complication was shown in the Table 5. The electrolyte imbalance which needed to be corrected was found to be higher in the Tertile-1 group. Other medical complications and surgical complication has no difference.

Table 5.

Postoperative Complication.

	Psoas Tertile 1 (N = 40)	Psoas Tertile 2 (N = 40)	Psoas Tertile 3 (N = 40)	P value
Post-operative medical complications
- Electrolyte imbalanaces	10 (66.70%%)	2 (13.30%)	3 (20%)	.01
- UTI	7 (33.30%)	9 (42.90%)	5 (23.80%)	.50
- Anemia	6 (35.30%)	5 (29.40%)	6 (35.30%)	.93
- Pneumonia	4 (66.70%)	1 (16.70%)	1 (16.70%)	.20
- Pulmonary embolism	0 (0%)	0 (0%)	1 (100%)	.37
- Heart complication	1 (2.4%)	0 (0%)	1 (2.6%)	.60
- Shock	0 (0%)	3 (75%)	1 (25%)	.16
- UGIB	0 (0%)	1 (100%)	0 (0%)	.36
- DVT	1 (50%)	1 (50%)	0 (0%)	.60
- Surgical site infection	1 (20%)	3 (60%)	1 (20%)	.43
- Wound dehiscence	0 (0%)	1 (50%)	1 (50%)	.60
- Other	1 (50%)	1 (50%)	0 (0%)	.61
Surgical complications	2 (33.30%)	3 (50%)	1 (16.70%)	.60

Discussion

According to previous studies, sarcopenia is the one of the predictors of the mortality following spinal metastatic surgery, the PMA has become a surrogate of sarcopenia. Most of studies, the PMA was identified by cross-sectional view of computer tomography at L3 or L4 pedicle level. Bourassa-Moreau E. et al studied the sarcopenia for prediction the early mortality and adverse events in emergent spinal metastatic surgery, the result showed the sarcopenia predicted 3-month mortality, independent of primary tumor.¹⁷ Dohzono S. et al studied showed the lowest percentage quartile (0-25%) of the psoas muscle mass was associated with poor overall survival rate in cervical spine metastatic surgery (HR 1.93, 95% confident interval 1.12-3.32; P = .0017).¹⁸ Hu MH. et al. studied, PMA is the prognosticator for 90-day survival and improves the discriminatory ability of earlier-proposed PSS in Asian populations (c-index, .74; 95% CI .67-.82 vs c-index, .74; 95% CI .66-.81).¹⁵ Pielkenrood B.J. et al studied a body fat distribution and sarcopenia on the overall survival in spinal metastatic patients who received radiotherapy, showed the muscle density was statistically significant associated with survival after 90 and 365 days, but body fat distribution was not significantly associated with overall survival.¹⁹

MRI is generally considered to be a more sensitive and specific modality for evaluating soft tissue, such as the muscles in the abdominal and pelvic region. MRI can also provide detailed images of the spine and pelvis, which may be important in some cases where those information is being used to guide surgical planning. Additionally, MRI does not expose the patient to ionizing radiation, which may be a consideration in certain clinical situations. In some circumstances, patients require urgent spinal surgery, such as decompression of cord compression. The selected imaging modality should be MRI, as it provides more information for surgical planning.

This study investigated the relation between the PMA under magnetic resonance and overall mortality rate. From our knowledge, this is the first study that used magnetic resonance imaging to evaluate the PMA in spinal metastatic patients who underwent the palliative spinal surgery. The PMA in T1 and T2 had a tendency of high 90-d mortality rate when compared to PMA in T3, the multivariable regressions showed that patients with low PMA had 38% higher mortality rate than large PMA in overall time. Kaplan–Meier graph showed the relation between psoas tertiles and time of death; PMA in T3 (large PMA) had a better probability of mortality rate than PMA in T1 and 2 (low PMA). According to the ICC was .76 (95% CI, .66–.84), the correlation between PMA under magnetic resonance imaging and computer tomography has a good reliability, but our study has a very low rate of the populations with computer tomography of lumbar spine. The preoperative characteristic is one of the prognostic factors which has a probability to predict the mortality rate; multivariable regression for the comorbidity showed 59% higher mortality rate than non-comorbidity in populations. Although, our study has not statistically difference in psoas tertiles at 90-d, 1-year, and overall mortality but there was a tendency of high mortality in low psoas group (PMA in T1 and T2), especially in 90-d mortality. Most of spinal metastatic patients have been undergone the magnetic resonance of screening whole spine for diagnosis and treatment, so this the benefit to evaluate the PMA in this imaging.

Limitations

There are several limitations in this study. First, this study limited by the retrospective review, which can potentially create a hidden bias due to the possibility of incomplete or inaccurate data. The population is heterogeneous in term of clinical decision-making and postoperative therapy due to the improvement of modality and treatment between 2009 and 2020 with a relatively small number of patients. In demographic data, the comparison between the psoas tertiles turned out of be insignificant but most population have the thoracic and lumbar spinal metastatic surgery, there was low cervical spinal metastatic surgery which may have a higher mortality rate than other site of spine surgery.

Conclusion

The MRI-based PMA tend to predict the 90-d mortality rate and overall mortality rate in spinal metastasis patients who undergone the spinal surgery. The PMA could be one of prognostic factors which can be used in decision-making, cooperated with other prognostic factors, in treatment of spinal metastatic patients. However, the future large-sized, prospective, and preferably multicenter studies are required to confirm or deny our results.

Supplemental Material

Supplemental Material - The Magnetic Resonance Imaging of the Psoas Muscle Area as the Prognostic Factor for Survival and Adverse Events in Spinal Metastasis Surgery

Supplemental Material for The Magnetic Resonance Imaging of the Psoas Muscle Area as the Prognostic Factor for Survival and Adverse Events in Spinal Metastasis Surgery by Kitidate Boonchai, Borriwat Santipas, Sirichai Wilartratsami, Monchai Ruangchainikom, Ekkapoj Korwutthikulrangsri, Pawin Akkarawanit, and Panya Luksanapruksa in Global Spine Journal.

Footnotes

Acknowledgments

The authors gratefully acknowledge Miss Sara Rawdeng of the Division of Research, Department of Orthopaedic Surgery, Faculty of Medicine Siriraj Hospital, Mahidol University for assistance with data collection.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Borriwat Santipas

Sirichai Wilartratsami

Monchai Ruangchainikom

Panya Luksanapruksa

Supplemental Material

Supplemental material for this article is available online.

References

Choi

Koo

Choe

Kang

. The Incidence and Management Trends of Metastatic Spinal Tumors in South Korea: A Nationwide Population-based Study. Spine. 2020;45(14):E856-E863.

Parkin

Pisani

Ferlay

. Global cancer statistics. CA Cancer J Clin. 1999;49(1):33-64.

Ware

JE,

Jr.

SF-36 health survey update. Spine. 2000;25(24):3130-3139.

Park

Lee

Chung

Lee

. How accurately can tokuhashi score system predict survival in the current practice for spinal metastases?: prospective analysis of 145 consecutive patients between 2007 and 2013. J Spinal Disord Tech. 2015;28(4):E219-E224.

Lee

Chung

Jahng

, et al. Which one is a valuable surrogate for predicting survival between Tomita and Tokuhashi scores in patients with spinal metastases? A meta-analysis for diagnostic test accuracy and individual participant data analysis. J Neuro Oncol. 2015;123(2):267-275.

Hibberd

Quan

GMY

. Accuracy of Preoperative Scoring Systems for the Prognostication and Treatment of Patients with Spinal Metastases. Int Sch Res Notices. 2017;2017:1320684.

Balain

Jaiswal

Trivedi

Eisenstein

Kuiper

Jaffray

. The Oswestry Risk Index: an aid in the treatment of metastatic disease of the spine. Bone Joint Lett J. 2013;95-b(2):210-216.

Tokuhashi

Uei

Oshima

Ajiro

. Scoring system for prediction of metastatic spine tumor prognosis. World J Orthop. 2014;5(3):262-271.

Wibmer

Leithner

Hofmann

, et al. Survival analysis of 254 patients after manifestation of spinal metastases: evaluation of seven preoperative scoring systems. Spine. 2011;36(23):1977-1986.

10.

Dardic

Wibmer

Berghold

Stadlmueller

Froehlich

Leithner

. Evaluation of prognostic scoring systems for spinal metastases in 196 patients treated during 2005-2010. European Spine Journal: Official Publication of the European Spine Society, the European Spinal Deformity Society and the European Section of the Cervical Spine Research Society. 2015;24(10):2133-2141.

11.

Schoenfeld

Marjoua

, et al. Assessing the utility of a clinical prediction score regarding 30-day morbidity and mortality following metastatic spinal surgery: the New England Spinal Metastasis Score (NESMS). Spine J. 2016;16(4):482-490.

12.

Karhade

Ahmed

Pennington

, et al. External validation of the SORG 90-day and 1-year machine learning algorithms for survival in spinal metastatic disease. Spine J. 2020;20(1):14-21.

13.

Cruz-Jentoft

Bahat

Bauer

Writing Group for the European Working Group on Sarcopenia in Older People 2 EWGSOP2, and the Extended Group for EWGSOP2 , et al. Sarcopenia: revised European consensus on definition and diagnosis. Age Ageing. 2019;48(1):601-631.

14.

Dohzono

Sasaoka

Takamatsu

Hoshino

Nakamura

. Prognostic value of low psoas muscle mass in patients with cervical spine metastasis. J Clin Neurosci. 2019;66:56-60.

15.

Yen

Chen

, et al. Decreased psoas muscle area is a prognosticator for 90-day and 1-year survival in patients undergoing surgical treatment for spinal metastasis. Clin Nutr. 2022;41(3):620-629.

16.

Badhiwala

Leung

Jiang

, et al. In-hospital Course and Complications of Laminectomy Alone Versus Laminectomy Plus Instrumented Posterolateral Fusion for Lumbar Degenerative Spondylolisthesis: A Retrospective Analysis of 1804 Patients from the NSQIP Database. Spine. 2021;46(9):617-623.

17.

Koo

. A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. J Chiropr Med. 2016;15(2):155-163.

18.

Bourassa-Moreau

Versteeg

Moskven

, et al. Sarcopenia, but not frailty, predicts early mortality and adverse events after emergent surgery for metastatic disease of the spine. Spine J. 2020;20(1):22-31.

19.

Pielkenrood

van Urk

van der Velden

, et al. Impact of body fat distribution and sarcopenia on the overall survival in patients with spinal metastases receiving radiotherapy treatment: a prospective cohort study. Acta Oncol. 2020;59(3):291-297.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.45 MB