Blood gases findings in patients with rib fractures: a retrospective cross-sectional study

INTRODUCTION

After cancer and vascular disease, trauma is the third most prevalent cause of mortality in the world, accounting for 9-10% of deaths.^1,2 It is also the primary reason for emergency admission and the source of almost 16% of all medical costs worldwide.¹ With an incidence of 45-65%, chest trauma is one of the most frequent injuries suffered by poly-trauma patients. Mortality rates for this kind of trauma, which is typically brought on by a high-energy blunt force, can reach 60%, which make it one of the leading causes of death in all age groups. 15% of all trauma admissions is related to chest trauma.^3-5 Rib fracture is the most common injury associated with chest trauma with overall mortality rate about 10%.^6,7

For critically ill patients, an arterial blood gas (ABG) gives crucial information. In general, the base excess (BE) is a useful indicator of substantial injuries in trauma patients (BE less than −4).⁸ In trauma patients suspected of being at risk for significant injury based on initial history and physical examination, there is evidence to support routine blood gas investigation for base deficit and metabolic acidosis evaluation. However, the best way to decide whether to have an arterial or venous blood gas is to evaluate the patient’s while taking arterial oxygen saturation and tissue perfusion status into account.⁹ Although carbon dioxide elimination may be almost unaffected, a crushed chest injury implies significant violence to the thorax, which mostly cause lung damage. This results in a decrease in the amount of oxygen in the arterial blood. The arterial blood’s oxygen content will continue to decrease due to ongoing hemorrhage, lobular atelectasis brought on by secretions, hemothorax and/or pneumothorax, discomfort, and central respiratory depression brought on by an injury or analgesic medication. Finally, when lung failure approaches, acidosis will develop and the arterial blood’s carbon dioxide concentration will increase.^10,11

Numerous studies have demonstrated that having more fractured ribs and their locations are risk factors for respiratory failure and death.¹² In a study, in deciding which patients were admitted to the ward and which were admitted to the intensive care unit, ABG values and combined ABG values were statistically significant. Patients admitted to the ICU had respiratory acidosis, hypercarbia, and hypoxia.⁶

Since rib fracture is the most common injury associated with chest trauma, our study aimed to find the factors of rib fracture that is associated with significant abnormalities in ABG and VBG, which are found in studies more predictive for ICU admissions.

Objective

To determine the relationship between arterial and venous blood gas abnormalities and the severity and characteristics of rib fractures, assessing their value in the diagnosis and prognosis of rib fracture patients.

METHODS

Study design

This was a cross-sectional study that aimed to measure the ABG and VBG values in patients that are presented with any kind of rib fractures, and provides supportive study that helps in the laboratory diagnosis and prognosis of rib fracture. We used observation records using a checklist that was tested and validated by a team of experts.

Study area

The study was conducted in the Institute of King Abdullah medical complex, Jeddah, Saudi Arabia carried out from December 2016 to January 2022.

Study population

From December 2016 through January 2022, we examined 120 electronic records of patients with x-ray confirmed rib fractures at King Abdullah Medical Complex Hospital. We excluded 54 records, either due to concurrent pulmonary diseases or missing ABG/VBG data, leaving 66 patients eligible for our final analysis.

The sample size of 66 corresponds to the total number of eligible patients with documented ABG or VBG during the study period. A randomized sample of these adult patients (≥18 years) with x-ray-confirmed rib fractures and documented ABG or VBG was selected from the electronic records of King Abdullah Medical Complex. We conducted simple random sampling from the hospital’s electronic records to reduce selection bias and ensure representation of all eligible patients during the study period. Patients with other pulmonary injury and diseases were excluded.

Data collection

A validated data collection sheet designed using Excel used to assess the correlation between different blood gasses and ribs fractures. The sheet consists of 4 sections: 1^st section: demographic data included age and sex, hospital stay and date of admission. 2^nd section: ribs and lung assessment: number of broken ribs, affected side (right or left), the exact location (anterior, posterior, lateral, antro-lateral, postero-lateral) the type of fracture (displaced or non-displaced), flail chest, lung contusions, other injuries, mechanism of injury.

3^rd section: ABG findings (PH, PO₂, PCO₂, HCO₃, O₂ saturation, base excess and deficit, anion gap with and without potassium (K), ABG interval): each was taken at admission and post admission with specifying the duration between them.

4^th section: VBG findings: (PH, PO₂, PCO₂, HCO₃, O₂ saturation, base excess and deficit, anion gap with and without potassium (K), VBG interval): each was taken at admission and post admission with specifying the duration between them.

The readings were taken twice if available, once at admission and the second was 24 hours post admission.

Statistical analysis

In the present study, statistical analysis using “IBM SPSS statistics version 23.0” test the objectives was applied to evaluate and Shapiro-Wilk test was used to test the normality of the study samples. Shapiro-Wilk test was used to test the normality of the study samples. Descriptive statistics to calculate median and quartiles. Descriptive statistics to calculate median and quartiles. Mann-Whitney test was applied to test the significant difference between two groups. Kruskal-Wallis test was applied to test the significant difference between three or more groups. Wilcoxon test was applied to test the significant difference between two related groups. Pearson’s correlation was used to measure the strength and direction of association between two variables.

Pearson’s correlation test was used to explore the relationships between age, number of fractured ribs, and both arterial and venous blood gas parameters, recognizing that no multivariable adjustment was performed in this retrospective design. Missing data were minimal (<5% for any variable) and were addressed through complete-case analysis by excluding records with missing values. Since our study included all eligible patient records during the study period (census sampling), no sampling weights or stratification were applied. Finally, no subgroup or sensitivity analyses were conducted.

Ethical considerations

The current study was approved by the research ethics committee of King Abdullah medical complex hospital at Jeddah, approval number H-02-J-002. In our study we didn’t collects any participants personal identifiers to protect their confidentiality. Furthermore, the collected data was used for research purposes only.

RESULTS

There were total 66 patients included in our study with mean age of the study sample of 40.5±18.8. 56 (84.8%) of them were males while the rest 10 (15.2%) were females. Missing data were minimal (<5% for all variables) and were handled via complete-case analysis. Baseline characteristics are summarized in (Table 1).

Blood gases

Arterial blood gasses

A total of 21 (31.81%) patients had an ABG at admission, compared to only 10 (15.15%) that had an ABG post admission and 7 (10.60%) had both reading at and post admission, their median for at admission versus post admission is as follows: PH were (7.301±0.09 versus 7.276±0.196; p value 0.612), PO₂ (110±143.425 versus 111±158.919; p value 0.499), PCO₂ (45.1±11.807 versus 44.6±14.761; p value 0.31), HCO₃ (21.05±4.803 versus 19.9±6.355), O₂ saturation (97.6±22.436 versus 97.15±19.054; p value 0.499), Base excess and deficit (-4.7±4.742 versus -5.15±8.211; p value 0.499) and anion gap (17.15±7.854 versus 10.4±9.819; p value 0.248) (Table 2). The correlation between ABG findings at admission and patient’s demographic data showed that older patients had a higher PCO₂ (r=0.474; p value 0.035) and anion gap (p value 0.036). Sex differences showed no relevant correlation.

Venous blood gasses

From the total of 42 (63.63%) patients had an VBG at admission, only 13 (19.69%) had an VBG post admission, while 11 (16.67%) of them had both reading at and post admission, their median for at admission versus post admission were as follows: PH (7.354±0.077 versus 7.315±0.276; p value 0.093), PO₂ (33.3±13.118 versus 40.5±22.673; p value 0.333), PCO₂ (43.05±7.237 versus 44.7±11.886; p value 0.286), HCO₃ (22.55±2.825 versus 19.9±5.51), O₂ saturation (59.4±21.989 versus 60.4±18.785; p value 0.79), base excess and deficit (-1.4±3.628 versus -4.3±7.167; p value 0.041) and anion gap (15.585±4.05 versus 15.45±8.205; p value 0.625) (Table 2). The correlation between VBG findings at admission and patient’s demographic data found that there was a significant decrease of O₂ saturation (r=-0.422; p value 0.007) found in older age patients, while there wasn’t significant correlation with longer sex differences.

Rib fracture

Affected side

There were 25 (37.88%) of patients with left side involvement while right rib fractures accounts for 31 (46.97%), the bilateral rib fractures were found in 10 (15,15%) (Table 3). There was significant relation with VBG changes at admission with PH for left, right and both sides as follows: (7.383 versus 7.318 versus 7.350; p value =0.008), respectively, and the PCO₂ (40.550 versus 45.200 versus 42.150; p value =0.027), respectively. There was no significant association between the affected side with gender or age.

Number of broken ribs

All the 66 patients had a broken rib with (Median 3±2.366). Increased number of broken ribs were significantly related to different VBG changes at admission, such as Ph (r: -0.448, p value: 0.003), PCO₂ (r: 0.376, p value: 0.014) and HCO₃ (r: -0.321, p value: 0.038). Also ABG PO₂ at admission is significantly increased with the number of broken ribs (p value: 0.043).

Right rib fracture

There were a total of 41 patients with involvement of the right ribs, 31 (75.6%) of them had isolated right ribs fractures, while 10 (24.4%) had bilateral involvement of ribs fractures. The 2^nd, 3^rd, and the 6^th ribs were most commonly fractured in the right side with 18 (27.3%), 16 (24.2%), and 15 (22.7%), respectively (Table 4). Nondisplaced fracture showed a slight tendency over displaced with 25 (37.9%) versus 19 (28.8%) while 3 (4.5) samples were committed. Posterior involvement of the right ribs were the most common involvement followed by anterior involvement with 21 (31.8%) and 16 (24.2%), respectively. In the VBG values at admission, when 3^rd, 4^th and 6^th ribs were involved, it predicts lower O₂ saturation (median 64 versus 39; p value =0.025), (Median 64 versus 39; p value =0.013), and (median 66 versus 45; p value =0.032), respectively. Usually, 6^th and 7^th ribs were associated with more abnormal VBG values at admission; beside lower O₂ saturation, 6^th right rib was also associated with lower PH, higher CO₂ and higher anion gap (medians; 7.366 versus 7.255; p value: 0.001, 46 versus 51; p value: 0.022, 17.2 versus 18.55; p value: 0.008, respectively), while the 7^th right rib fracture mostly predict lower PH, lower HCO₂ and higher anion gap (medians; 7.361 versus 7.25; p value: 0.009, 22.75 versus 19.85; p value: 0.04, 15.2 versus 18.8; p value: 0.039). The 12^th right rib also predicts higher anion gap (median: 15.3 versus 20.2; p value: 0.044). While in ABG 10^th right rib fracture involvement was associated with benign finding in comparison when it was not fractured (median; 7.237 versus 7.357; p value: 0.044).

Left rib fracture

There were a total of 35 patients with involvement of the left ribs, 25 (71.43%) of them had isolated left ribs fractures, while 10 (28.57%) had bilateral involvement of ribs fractures. The 6^th, 7^th, and 8^th were the most commonly involved in the left side with 13 (19.7%), 10 (15.2%), 10 (15.2%), respectively (Table 5). With 21 (31.8%) versus 18 (27.3%) for displaced versus nondisplaced in contrast to the right side. But the results regarding the exact location upon the rib were consistent with the right with 21 (31.8%), and 9 (13.6%) for posterior followed by anterior involvement. The relation to blood gasses at admission, the 6^th rib fractured predicted a higher VBG- HCO₃ (median 21.500 versus 23.600; p value =0.031), while the 7^th rib fractured had a lower ABG anion gap (medians; 18.850 versus 7.850; p value =0.028). There was a tendency for the left 7^th, 8^th,10^th ribs to be associated with lung contusion (2.9% versus 29%; p value =0.004), (2.9% versus 29%; p value =0.004), (22.9% versus 58.1%; p value =0.005) respectively.

Mechanism of injury

Regarding the mechanism of injury it shows variation but the major one which accounts for 57.6% was RTA, the next was fall with a total of 27.3 divided into 16.7% simple falling, while falling to another level was 10.6%. There was no association between the mechanism of injury and side affected (Table 6).

Other injuries

Regarding injury associated with the rib fractures; pneumothorax followed by hemothorax were the commonest injuries with incidence of 39.4% and 18.2%, respectively. The incidence of clavicular fracture, pleural effusion, scapular fracture, sternal fracture, and atelectasis were 12.1%, 12.1%, 10.6%, 9.1%, and 7.6%, respectively. Hemothorax and surgical emphysema were more likely to happen in older ages (median ages for their presence versus absence: 56 versus 31; p value: 0.008, and 71 versus 35; p value: 0.048).

Hospital stay

The mean of hospital stay was mean =12.485±20.941. There was no significant difference between males or females in terms of hospital stay, with a median of 6 versus 4, respectively; p value =0.184. There was a significant increase of hospital stay with ABG PO₂ values at admission; R=0.637, p value =0.00.

DISCUSSION

In this retrospective cross-sectional study, we demonstrated that increasing age and number of rib fractures were significantly associated with elevated PaCO₂ and anion gap in arterial blood gases (ABG) as well as decreased O₂ saturation in venous blood gases (VBG). These findings highlight the need for early respiratory monitoring in older patients and those with multiple rib fractures.

Blood gases

Rotab et al study showed that patients with chest trauma exhibit initial low arterial pH when they don’t have lung injuries compared to chest trauma patients with lung injury and when hospital stay was 4 days or more compared to less than 4 days (7.33±0.04 versus 7.35±0.09, and 7.32±0.13 versus 7.37±0.045, respectively), the study also showed that chest trauma patients with lung injuries show more increased arterial PCO₂ level compared to chest trauma patients with lung injury (43.23±9.43 versus 42.85±1.26), and decreased level of arterial PO₂, HCO₃, and SaO₂ (56.85±4.91 versus 62.84±2.50, 20.43±2.41 versus 23.37±2.32, and 81.75±17.38 versus 86.41±12.62, respectively).⁶ When the hospital stay was 4 days or more the study showed high arterial PCO₂ and HCO₂ (47.69±6.33 versus 40.27±5.59 and 23.77±2.72 versus 22.97±2.15, respectively) and low arterial PO₂ and SaO₂ (51.09±9.53 versus 68.79±8.47 and 77.27±13.48 versus 91.04±9.38). These findings are consistent with our finding regarding the low pH and high PCO₂, while our study contradicts it that HCO₃, and O₂ saturation were normal initially, knowing that HCO₃ did decrease after 24 hours, and high PO₂. While the result of Carlino et al showed similar results regarding non lung injury versus lung injury patients in arterial PaO₂ and SaO₂ (78.78±8.01 versus 64.45±11.46 and 98.11±0.78 versus 94.62±4.18, respectively) there was contradiction with the results of initial pH value for non lung injury versus lung injury which exhibit that pH in non lung injury was higher compared to lung injury group (7.44±0.03 versus 7.43±0.04).¹³ Tsai et al study calculated the following arterial results of PH, PaO₂, PaCO₂, HCO₃, and SaO₂ for survivors with the following results (7.33±0.06, 129.09±58.76, 40.89±8.10, 21.29±4.58, and 95.48±8.29, respectively) and for non-survivors the results were (7.28±0.133, 125.60±130.81, 45.01±10.99, 20.43±3.35, and 81.41±28.30, respectively), results which show increased levels of PaO₂ in contrast to Rotab et al and Carlino et al and consistent with our research results.^6,13,14 Bellone et al study only measured arterial SaO₂ parameter and calculated the mean with 96% (94-98%) in patients with blunt chest trauma.¹⁵ Measuring the difference in at admission ABG compared to 6 hours post admission Mohsenian et al found that PH, SaO₂, PCO₂, HCO₃, base excess at admission and 6 hours after admission in traumatic injuries with femur and pelvis fractured were 7.34±0.088 versus 7.32±0.086, 92.1±6.68 versus 95.98±2.47, 40.69±11.02 versus 37.99±10.44, 22.12±4.18 versus 20.40±4.05, and -3.251±4.117 versus - 4.531±5.748, respectively.¹⁶ Our did show decreased pH arterial value 24 hours later, with both value below normal range, our study showed an increased O₂ saturation and further increasing 24 hours later, PCO₂ in our study were both in normal ranges with decreasing in the PCO₂ value 24 hours later, the HCO₃ values did consistent with our findings except that we found an abnormal decreasing in the value 24 hours later, and regarding base deficit the finding were compatible.

Rib fracture

In Carlino et al study the number of broken ribs with a blunt chest detected by CT was higher in patients with lung injury in comparison to patients without lung injury (3.67±1.87 versus 1.33±1.23).¹³ Schmoekel et al study ribs fractures in patients aged 55 or older and found that patients having more than >6 ribs fractured were 18.3%, 13.4% having bilateral ribs involvement, and 7.6% having the first rib fractured.¹⁷ The number of fractured ribs in chest blunt trauma patients admitted to ICU in Bellone et al study were 3 ribs or more in 70% of cases and 30% were having less than 3 ribs involved with 4 ribs as a median for number of fractured ribs.¹⁵ In Shulzhenko et al the percentages for 1 fracture, 2 fractures, 3 fractures, 4 fractures, 5 fractures, 6 fractures, 7 fractures, and 8 fractures of the ribs were 23.1%, 18.3%, 15.7%, 15.4%, 8.6%, 6.0, 4.2%, and 8.8%, respectively.¹⁵ While in our study, the second rib was most commonly involved in the right chest while the sixth rib was the most commonly involved in the left side. Finally, 21% of patients with blunt trauma had rib fractures in Cameron et al study.¹⁸

Mechanism of injury

Carlino et al mentioned fall-related injury as the most common cause of chest trauma 64.7%, followed by motor vehicle crash 27.5%, findings which correlate with Shulzhenko et al study for the falls being the commonest mechanism of injury (43.2%) followed by motor vehicle collision (30.4%).^12,13 Tsai et al mentioned different findings with road accidents being the most common cause of chest injury followed by falling injury 66.67% versus 20%.¹⁴ Cameron et al agreed with the latter findings with the commonest cause for injury with 48.05% of ribs being fractured by motor vehicle collision.¹⁸ In our study motor vehicle collision was also the most common, followed by falling.

Other injuries

Hemothorax and pneumothorax were common to be reported as the common associated injury in chest trauma, Rotab et al and Tsai et al found hemothorax as the most common (60% in Tsai et al after rib fracture) while Carlino et al found the opposite, so was our findings.^13,14,16 The results of Carlino et al and our results were similar to schmoekel et al with pneumothorax being the commonest associated injury in elderly (23.6%) followed by hemothorax (10.6%).^13,17

Hospital stay

The length of hospital stay in Rotab et al study was 5.03±3.56 with a median of 4 days of hospital stay.⁶ Shulzhenko et al and Bellone et al found relatively longer mean of 7.9 and 6.4 days, respectively, while Cameron et al trial showed a longer duration in comparison with average length of hospital stay for rib fractures 16.16±24.3 days.^12,15,18 Tsai et al study found even a longer mean of hospital stay with 50.8 days.¹⁴ Our study found that the mean of hospital stay was 12.4 days.

We faced different limitations in our study, first is that this is a retrospective study, so it all depends on the registered data. Most of the patients who have had an ABG or VBG at admission don’t have post admission testing. Second is that the sample size is limited because of that not all the patients with blunt chest trauma and rib fracture received an ABG or VBG test. Third is that this study was done only in one center.

CONCLUSION

In our retrospective cross-sectional study, we observed that older patients and those with multiple rib fractures tended to have more pronounced abnormalities in both arterial and venous blood gas values. We believe this underscores the importance of closer respiratory monitoring in these high-risk patients. Although our single-center, retrospective design may limit generalizability, we nevertheless feel these findings are applicable to similar tertiary-care hospitals treating rib trauma and hope they will inspire larger prospective, multicenter studies to confirm and build upon our work. Our findings indicate that ABG and VBG values can be helpful in the management of rib fractures. Generally, patients exhibited low arterial and venous pH, which further decrease 24 hours later, normal PCO₂ values for arterial and venous samples at and post admission, except for slight elevation in arterial PCO₂ at admission, normal at admission arterial and venous HCO₃, which got into a slight abnormal decreasing in post admission values, decreased base excess which also had a further decreasing in post admission values. Median of 3 ribs fractured for a patient, with pneumothorax being the most common associated injury followed by hemothorax. Motor vehicle collision is the most common mechanism of injury followed by falling, mean hospital stay was 12.485 for a patient.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: The study was approved by the Institutional Ethics Committee of King Abdullah medical complex hospital at Jeddah, approval number H-02-J-002

REFERENCES

1. 1. Akbaryan F, Derakhshanfar H, Heidari K, Vafaei A, Matin S. Prognosis in children with traumatic injuries referring to the emergency department. Chin J Traumatol. 2019;22:323-7.
2. 2. Magu S, Yadav A, Agarwal S. Computed tomography in blunt chest trauma. Indian J Chest Dis Allied Sci. 2009;51:75-81.
3. 3. Ried M, Bein T, Philipp A, Müller T, Graf B, Schmid C, et al. Extracorporeal lung support in trauma patients with severe chest injury and acute lung failure: a 10-year institutional experience. Crit Care. 2013;17:R110.
4. 4. Hunt PA, Greaves I, Owens WA. Emergency thoracotomy in thoracic trauma- a review. Injury. 2006;37:1-19.
5. 5. Cameron P, Dziukas L, Hadj A, Clark P. Rib fractures in major trauma. Aust N Z J Surg. 1996;66:530-4.
6. 6. Rotab OA, Gwely NE, Atia SM, Ismael HK. Arterial blood gases as an indicator of early ICU admission in chest trauma patients. Egypt J Hosp Med. 2021;82:471-8.
7. 7. Coary R, Skerritt C, Carey A, Rudd S, Shipway D. New horizons in rib fracture management in the older adult. Age Ageing. 2020;49:161-7.
8. 8. Rutherford EJ, Morris JA Jr, Reed GW, Hall KA. Base deficit stratifies mortality and determines therapy. J Trauma. 1992;33:417-23.
9. 9. Jousi M, Reitala J, Lund V, Katila A, Leppäniemi A. The role of pre-hospital blood gas analysis in trauma resuscitation. World J Emerg Surg. 2010;5(1):10.
10. 10. Harley HR. The causes of respiratory failure in closed crush injuries of the chest and upper abdomen. Proc R Soc Med. 1961;54:558-61.
11. 11. Ganie FA, Lone H, Lone GN, Wani ML, Singh S, Dar AM, et al. Lung contusion: a clinico-pathological entity with unpredictable clinical course. Bull Emerg Trauma. 2013;1:7-16.
12. 12. Shulzhenko NO, Zens TJ, Beems MV, Jung HS, O’Rourke AP, Liepert AE, et al. Number of rib fractures thresholds independently predict worse outcomes in older patients with blunt trauma. Surgery. 2017;161:1083-9.
13. 13. Carlino MV, Guarino M, Izzo A, Carbone D, Arnone MI, Mancusi C, et al. Arterial blood gas analysis utility in predicting lung injury in blunt chest trauma. Respir Physiol Neurobiol. 2020;274:103363.
14. 14. Tsai YM, Lin KH, Huang TW, Chen CY, Hong ZJ, Hsu SD. Outcomes of patients with blunt chest trauma encountered at emergency department and possible risk factors affecting mortality. J Med Sci. 2017;37:97-101.
15. 15. Bellone A, Bossi I, Etteri M, Cantaluppi F, Pina P, Guanziroli M, et al. Factors associated with ICU admission following blunt chest trauma. Can Respir J. 2016;2016:3257846.
16. 16. Mohsenian L, Khoramian MK, Sadat Mazloom S. Prognostic value of arterial blood gas indices regarding the severity of traumatic injury and fractures of the femur and pelvis. Bull Emerg Trauma. 2018;6:318-24.
17. 17. Schmoekel N, Berguson J, Stassinopoulos J, Karamanos E, Patton J, Johnson JL. Rib fractures in the elderly: physiology trumps anatomy. Trauma Surg Acute Care Open. 2019;4:e000257.
18. 18. Cameron P, Dziukas L, Hadj A, Clark P, Hooper S. Rib fractures in major trauma. Aust N Z J Surg. 1996;66:530.