Predicting pulmonary complications and cost burden in Saudi orthopedic patients: A prospective modified ARISCAT score validation

Abstract

Background:

Postoperative pulmonary complications (PPCs) are a major cause of morbidity in orthopedic surgery, yet the Assess Respiratory risk in Surgical Patients in Catalonia (ARISCAT) score remains under-validated in Saudi populations with unique metabolic and environmental risk profiles.

Methods:

We conducted a prospective cohort study (2024–2025) of 600 adults undergoing elective/urgent orthopedic surgery at King Abdulaziz Medical City. Using a prespecified +5-point adjustment for cemented arthroplasty (accounting for bone cement implantation syndrome), we calculated ARISCAT scores preoperatively and assessed PPCs within 7 days using European Society of Anesthesiology and Intensive Care criteria. Secondary outcomes included intensive care unit admissions, length of stay (LOS), and hospitalization costs.

Results:

The overall PPC incidence was 19.2%, with stepwise increases by risk category: 4.1% (low risk, ≤25 points), 17.3% (intermediate), and 44.8% (high-risk, ≥45; P < 0.001). The model demonstrated excellent discrimination (area under the curve [AUC] = 0.83, 95% confidence interval [CI] 0.78–0.88) and calibration (P = 0.27). High-risk patients had 5.2× greater PPC odds (adjusted odds ratio [OR] 5.2, 95% CI: 3.1–8.7), 3.1-day longer LOS, and $2350 higher costs versus low-risk patients (all P < 0.001). Cemented arthroplasty independently predicted PPCs (OR 1.8, 95% CI: 1.1–3.0).

Conclusion:

This prospective Saudi validation demonstrates ARISCAT’s strong predictive performance in orthopedic surgery, with notable enhancement when applying the cemented procedure modification. The substantial clinical burden observed in high-risk patients (44.8% PPC rate) and the associated economic impact highlight an urgent need for targeted preoperative optimization and resource-stratified care pathways. These findings reinforce ARISCAT’s utility in guiding surgical planning and underscore the value of tailoring risk assessment tools to the unique metabolic and environmental risk profile of Saudi orthopedic populations.

Background

Postoperative pulmonary complications (PPCs) have emerged as a critical determinant of surgical outcomes in orthopedic populations, particularly in regions with unique demographic and environmental risk profiles.[1] In Saudi Arabia, PPCs account for 38% of preventable postoperative deaths according to the 2023 National Surgical Outcomes Report,[2] a rate 1.9-fold higher than global averages reported by the European Perioperative Clinical Outcomes network.[3] This disparity reflects the intersection of three synergistic risk factors that characterize the Kingdom’s surgical population. First, the high prevalence of metabolic syndrome components–including 39.7% obesity and 34.2% diabetes among orthopedic patients[4]–has been mechanistically linked to impaired respiratory function through multiple pathways. A 2022 study at King Fahad Medical Center demonstrated these metabolic factors independently increase PPC risk by 2.3-fold (95% CI: 1.8–3.1) through combined effects on diaphragmatic function, pulmonary compliance, and immune response.[5]

Second, Saudi Arabia’s distinctive environmental conditions substantially elevate baseline pulmonary risk. The Red Sea coastal region experiences >56 annual sandstorm days,[6] with particulate matter (PM10) concentrations exceeding 300 μg/m³ during events.[7] Multicenter research led by Al-Otair et al. established that preoperative sandstorm exposure doubles PPC odds (OR 2.1, 95% CI: 1.3–3.4) through direct epithelial injury and inflammatory cascades.[8] Third, procedural patterns differ markedly from Western cohorts, with cemented arthroplasties comprising 53% of joint replacements versus 35% in North American registries.[9] This is particularly consequential given the 22.5% PPC incidence in cemented cases versus 13.5% in cementless procedures (P < 0.001) documented in a 2023 analysis of 1200 Saudi patients.[10]

The Assess Respiratory risk in Surgical Patients in Catalonia (ARISCAT) risk score, while validated across European and American populations with area under the curves (AUCs) of 0.80-0.89,[11,12] demonstrates critical limitations in this context. Derived from cohorts with substantially lower metabolic risk (<17% obesity prevalence[13]), the tool lacks weighting for sandstorm exposure or cemented procedures. Furthermore, retrospective Saudi data show 32% misclassification of respiratory complications when using standard diagnostic codes,[14] while economic analyses reveal PPCs increase hospitalization costs by $2350 per case.[15]

Our prospective study addresses these gaps through four innovations that advance perioperative risk assessment: (1) population-specific validation across 600 consecutive patients at King AbdulAziz Medical City, a tertiary center, (2) protocolized PPC adjudication using European Society of Anesthesiology and Intensive Care (ESAIC) criteria,[16] (3) a +5-point cemented procedure modification (validated by OR 1.8, 95% CI: 1.1–3.0 in our pilot data[17]), and (4) detailed cost modeling projecting $3.2 million annual savings through targeted prevention.[18] These developments promise to transform risk stratification in a population where current tools may underestimate true complication risks by 15%–20%.[19]

Methods

Study design and setting

This prospective observational cohort study adhered to STROBE guidelines and was conducted at King Abdulaziz Medical City (KAMC), Jeddah (January 2024–June 2025). Institutional Review Board approval was obtained (KAIMRC), with waived informed consent per Saudi FDA ethical guidelines. Data was obtained from the hospital’s BestCare electronic medical record system. The use of routinely collected clinical data maintained the study’s noninterventional design, while allowing comprehensive capture of patient demographics, comorbidities, intraoperative variables, and postoperative outcomes.

Participants

Consecutive adult patients (≥18 years) undergoing elective, urgent, or emergent orthopedic surgeries under general/regional anesthesia were enrolled.

Key definitions

• Urgent surgery: Required within 48 hours (e.g., infected arthroplasty) but not immediately life-threatening (<6 hours = emergent)[3]

• Exclusions: Trauma-related procedures (due to heterogeneous injury patterns[4]), pre-existing ventilation, incomplete data, or surgeries <60 min.[5]

Assess respiratory risk in surgical patients in Catalonia score and modifications

We calculated the ARISCAT score using its seven original components (age, SpO₂, respiratory infection, hemoglobin, surgery duration, surgical site, emergency status) [Table 1],[5] with modifications:

Table 1.

Parameters of the Assess Respiratory risk in Surgical Patients in Catalonia Score and risk classification

Score components	Risk score
Age (years)
≤50	0
51–80	3
>80	16
Preoperative oxygen saturation (%)
≥96	0
91–95	8
≤90	24
Respiratory infection in past 1 month
No	0
Yes	17
Preoperative hemoglobin <10
No	0
Yes	11
Incision
Peripheral incision	0
Upper abdominal incision	15
Intrathoracic incision	24
Surgical duration (h)
<2	0
2–3	16
>3	23
Emergency procedure
No	0
Yes	8
Risk	ARISCAT score
Low	<26
Medium/intermediate	26–44
High	≥45

1. +5 points for cemented arthroplasty (validated by OR 1.8 for PPCs[6])

2. SpO₂ measurement: After 5 minutes of rest (WHO protocol[7]).

Outcomes

• Primary: PPCs within 7 days (ESAIC criteria[8]): Pneumonia, respiratory failure, atelectasis, bronchospasm, or unplanned ventilation >24 h

• Secondary: Intensive care unit (ICU) admissions, length of stay (LOS), costs (inflation-adjusted[9]).

Statistical analysis

• Descriptive: Mean ± standard deviation or median (interquartile range)

• Comparisons: ANOVA/Kruskal–Wallis (normality via Shapiro–Wilk), Chi-square/Fisher’s exact test

• Model performance: AUC-receiver operating characteristic (ROC) (95% CI: AUC >0.75 = clinically useful[12]), Hosmer-Lemeshow test[13]

• Multivariable analysis: Adjusted for age, diabetes, anesthesia type, body mass index (BMI).[14]

Results

The study cohort demonstrated striking physiological and procedural gradients across ARISCAT risk strata [Table 2]. High-risk patients (≥45 points) were significantly older (mean age 68.2 vs. 52.1 years, P < 0.001), with higher rates of hypoxemia (SpO₂ <90%: 27% vs. 2.6%) and cemented arthroplasties (74.3% vs. 16.7%) compared to low-risk counterparts. These baseline differences translated to clinically meaningful outcome disparities, as detailed below:

Table 2.

Baseline characteristics and procedure-specific outcomes by Assess Respiratory risk in Surgical Patients in Catalonia risk group

Variable	Total (n=600)	Low risk (≤25)	Intermediate risk (26–44)	High risk (≥45)	*P*
Demographics
Age (years), mean±SD	59.3±12.7	52.1±10.2	60.5±11.8	68.2±9.5	<0.001*
Female, n (%)	336 (56)	112 (57)	148 (57)	76 (52)	0.62
Comorbidities, n (%)
Diabetes	205 (34.2)	35 (18)	110 (42)	60 (41)	<0.001*
SpO2 <90%*	74 (12.4)	5 (2.6)	30 (11.5)	39 (27)	<0.001*
Procedures
Joint arthroplasty	408 (68)	60 (31)	200 (77)	148 (82)	<0.001*
Cemented, n/N (%)	220/408 (53.9)	10/60 (16.7)	100/200 (50)	110/148 (74.3)	<0.001*
PPC rate, n/N (%)	92/408 (22.5)	2/60 (3.3)	40/200 (20)	50/148 (33.8)	<0.001*
Spinal procedures	192 (32)	135 (69)	60 (23)	7 (18)	<0.001*
PPC rate, n/N (%)	26/192 (13.5)	6/135 (4.4)	15/60 (25)	5/7 (71.4)	<0.001*
Urgent surgery, n (%)	132 (22)	15 (8)	60 (23)	57 (39)	<0.001*

*SpO₂ <90%=Preoperative oxygen saturation <90% on room air. PPC=Postoperative pulmonary complication, SD=Standard deviation

Risk stratification and postoperative pulmonary complication incidence

The distribution of PPC subtypes across ARISCAT risk categories showed a clear stepwise increase in complication rates, from 4.1% in the low-risk group to 44.8% in the high-risk group. Pneumonia and respiratory failure were the most frequent complications in high-risk patients, as presented in Figure 1.

Figure 1.

Postoperative pulmonary complications subtype distribution

Model performance

Figure 2 presents the ROC curve assessing the ARISCAT score’s predictive performance for PPCs. The model demonstrated excellent discrimination with an AUC of 0.83 (95% CI: 0.78–0.88), maintaining strong predictive capability across both cemented (AUC 0.81) and noncemented (AUC 0.85) arthroplasty subgroups. The curve’s trajectory shows optimal sensitivity-specificity balance at the ≥45 cutoff (84% and 79%, respectively), supporting its clinical utility for risk stratification in our orthopedic cohort as demonstrated in Figure 2.

Figure 2.

Receiver operating characteristic curve for Assess Respiratory Risk in Surgical Patients in Catalonia performance

Multivariable analysis of postoperative pulmonary complication predictors

After adjustment for age, diabetes, anesthesia type, and BMI, multivariable analysis identified three independent predictors of PPCs: preoperative hypoxemia (SpO₂ <90%; OR 3.1, 95% CI: 1.7–5.6), recent respiratory infections (OR 4.6, 95% CI: 2.2–9.3), and cemented arthroplasty (OR 1.8, 95% CI: 1.1–3.0). Prolonged surgery duration (>3h) showed borderline significance (OR 2.2, 95% CI: 1.3–3.9), while age >70 years was nonsignificant [Table 3].

Table 3.

Multivariable logistic regression results

Predictor	Adjusted OR (95% CI)	P
SpO2 <90%	3.1 (1.7–5.6)	<0.001
Recent respiratory infection	4.6 (2.2–9.3)	<0.001
Surgery duration >3h	2.2 (1.3–3.9)	0.007
Cemented arthroplasty	1.8 (1.1–3.0)	0.02
Age >70 years	1.5 (0.9–2.4)	0.12

OR=Odds ratio, CI=Confidence interval

Economic and clinical impact

Using hospital-derived data on LOS, ICU admissions, and procedural risk groups, we modeled the potential economic burden of PPCs and the expected savings from ARISCAT-guided interventions. The actual LOS and ICU admission rates in our cohort were obtained from the hospital system, while cost figures and PPC reduction estimates were extrapolated from published economic analyses in similar settings.

In our cohort, high-risk patients (ARISCAT ≥45) accounted for 44.8% of PPCs (50/148 cases), 28% of ICU admissions (40/145), and had a mean LOS 2.5-fold longer than low-risk patients (7.1 ± 3.0 vs. 2.8 ± 0.9 days, P < 0.001). Costs were modeled based on published per-day hospitalization and ICU rates in Saudi Arabia, adjusted to our observed LOS distribution. This yielded an estimated mean cost of $4300 ± $1500 for high-risk patients, versus $1200 ± $400 for low-risk patients [Table 4].

Table 4.

Economic outcomes (length of stay, costs) by risk group.

Outcome	Low risk	Intermediate risk	High risk	P
Mean LOS (days)	2.8±0.9	4.5±1.7	7.1±3.0	<0.001
Mean cost (USD)	$1200±$400	$2600±$900	$4300±$1500	<0.001
ICU admissions, n (%)	2 (1)	18 (7)	40 (28)	<0.001

*Costs modeled from length of stay using published Saudi cost estimates forward and ICU stays. LOS=Length of stay, ICU=Intensive care unit

Projections based on these modeled estimates suggest that a 1% reduction in PPC rate could yield approximately $38,500 in annual savings per 1000 surgeries. Universal ARISCAT implementation could potentially prevent 22–30 PPCs annually at our institution, corresponding to an estimated $85,000–$116,000 in savings, if paired with targeted measures such as preoperative optimization and postoperative high-dependency monitoring for high-risk patients. Real-world savings would depend on actual compliance, local resource costs, and patient mix.

These modeled findings highlight both the clinical and economic relevance of ARISCAT-guided risk stratification, particularly in resource-conscious settings where high-risk patients (≥45 points) drive most preventable costs through prolonged LOS and frequent ICU utilization [Figure 3].

Figure 3.

Length of stay versus Assess Respiratory Risk in Surgical Patients in Catalonia score correlation

Figure 3 represents a Scatter plot of LOS versus ARISCAT score, stratified by PPC status and risk tier. The cost estimates associated with each LOS value were modeled using published Saudi cost data for forward and ICU admissions. High-risk patients (≥45 points) showed both longer LOS and higher ICU utilization, accounting for the majority of modeled preventable costs.

Discussion

This study represents one of the first prospective validations of the ARISCAT score in a Saudi orthopedic cohort, addressing critical gaps regarding its applicability to populations with high metabolic and environmental risk factors.[3,4] The model demonstrated strong predictive accuracy (AUC 0.83), comparable to its original European validation (AUC 0.89),[2] despite our cohort’s markedly higher prevalence of obesity (39.7% vs. 16.9%) and diabetes (34.2% vs. 8%–12%).[13] This robustness suggests the ARISCAT score’s generalizability across populations, although our +5-point modification for cemented arthroplasty—motivated by regional surgical trends and the documented risk of bone cement implantation syndrome[10]—may enhance predictive precision in similar settings.

Three findings are of particular clinical and economic importance. First, modifiable preoperative factors emerged as strong predictors: oxygen saturation <90% (OR 3.1) and recent respiratory infections (OR 4.6), reinforcing ESAIC recommendations for preoperative optimization.[6] In Saudi Arabia, where sandstorms[4] and elevated smoking rates (>14%[19]) impair respiratory health,[4] targeted measures such as delaying surgery in the presence of active infection or implementing preoperative bronchodilator therapy could potentially reduce PPC incidence by 18%–22%, consistent with data from comparable populations.[14]

Second, cemented arthroplasty was associated with a 1.8-fold higher risk of PPCs, validating our modified ARISCAT score’s additional weighting for this surgical technique. This finding supports perioperative adjustments such as refinements in cementing technique and vigilant intraoperative hemodynamic monitoring.[11]

Third, modeling based on our cohort data suggests that PPCs could potentially double hospitalization stays (projected: 6.2 vs. 3.1 days) and increase costs by approximately 2.5-fold (estimated: $3850 vs. $1500 per case) in comparable patient populations. If these patterns hold at scale, preventing just 10 PPC cases annually might generate sufficient savings (~$38,500) to theoretically fund hospital-wide preventive interventions like incentive spirometry programs,[15] though actual savings would depend on local care pathways and resource pricing structures.

While the prospective design reduces biases common in retrospective analyses,[4] the single-center scope may limit external generalizability. Nonetheless, the sample size (n = 600) and the use of standardized ESAIC PPC definitions[6] strengthen the internal validity of the results. The absence of long-term pulmonary outcome data is a limitation, given that 20–30% of PPC survivors develop persistent impairment.[16] Future studies should incorporate a 6- to 12-month follow-up to assess the ARISCAT score’s utility in predicting chronic sequelae.

Conclusion

This prospective validation confirms the ARISCAT score’s utility for predicting PPCs in Saudi orthopedic patients, even in high-metabolic-risk populations. The significant impact of metabolic comorbidities on pulmonary outcomes has been previously highlighted in the Saudi context,[20] and our results reinforce the critical need to address these factors preoperatively. The model’s strong discrimination (AUC 0.83), cost-saving potential (~$38,500/year per 1% PPC reduction), and adaptability to cemented procedures support its immediate clinical adoption.

To translate these findings into practice, we recommend:

1. National implementation with local adaptations: Integrate ARISCAT into preoperative clinics across Saudi hospitals, with mandatory SpO₂ screening and respiratory infection assessments.[6,14] Retain the +5-point adjustment for cemented arthroplasty (validated by OR 1.8 for PPCs), but validate this modification in trauma populations

2. Targeted preoperative optimization: Delay elective surgery for patients with active respiratory infections (OR 4.6) or SpO₂ <90% (OR 3.1), pairing this with 1–2 weeks of preoperative inspiratory muscle training (shown to reduce PPCs by 40% in high-risk groups). Screen for sandstorms and smoking-related respiratory exacerbations in arid regions using region-specific questionnaires

3. Resource-stratified monitoring: For high-risk patients (ARISCAT ≥45), mandate postoperative high-dependency unit monitoring for 24 h to reduce ICU overutilization. For intermediate-risk patients (26–44), implement protocolized incentive spirometry and early mobilization to cut LOS by 1.5 days

4. Policy considerations: If validated in broader populations, these findings could inform two key initiatives:

   1. Development of national perioperative guidelines incorporating ARISCAT risk stratification, potentially aligned with ESAIC standards. Such guidelines might prioritize high-risk patients (≥45 points) for targeted interventions

   2. Pilot testing of cost-sharing models for preoperative optimization (e.g., physiotherapy), where projected savings of ~$3850 per avoided PPC could theoretically offset program costs. Implementation would require assessment of local cost structures and stakeholder buy-in.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.