Introduction

Since the average life expectancy has increased, the majority of patients undergoing hip arthroplasty are elderly. The key to the recovery of elderly patients includes early functional training and a reduction in postoperative complications, and regional anaesthesia could achieve these goals1,2. The innervation of the hip joint includes the femoral, obturator, sciatic, and superior gluteal nerves, which can be blocked by a lumbar and sacral plexus block (LSPB)2. Karaca et al. 3 retrospectively analysed the mortality rates of 257 hip fracture patients and reported that LSPB resulted in less haemodynamic disturbance and greater cardiovascular stability. Moreover, lumbar plexus block has been reported to significantly reduce opioid consumption and blood loss when used to supplement general anaesthesia (GA) for hip arthroplasty4. Therefore, the combination of GA with LSPB provides optimal anaesthesia for hip arthroplasty.

Postinduction hypotension (PIH) frequently manifests in the interim between anaesthetic induction and skin incision. It is caused by decreases in cardiac output (CO), heart rate (HR) and systemic vascular resistance (SVR)5, especially in elderly patients, due to cardiovascular dysfunction and volume deficiency6,7. Severe episodes of intraoperative hypotension are associated with prolonged hospital stays and increased one-year mortality rates8. Although LSPB involves more limited sympathectomy than SA does, Visme et al.9 reported that the decrease in mean arterial pressure (MAP) from baseline was 27% in elderly patients after LSPB. Hence, there is increasing concern about the development of hypotension in elderly patients due to the dual negative effects on sympathetic activation produced by GA combined with LSPB, making predicting PIH an important issue in elderly patients undergoing hip arthroplasty.

Ultrasound measurement of the inferior vena cava collapsibility index (IVCCI) and inferior vena cava (IVC) diameter has the advantages of easy performance, repeatability and low cost and has often been used to predict fluid responsiveness in emergency and critical care settings10,11,12. Insufficient fluid volume is considered when the IVCCI increases or the diameter of the IVC decreases11. Recently, this noninvasive haemodynamic assessment method was introduced to predict PIH perioperatively in the operating room. However, its utility remains controversial across different population settings and different types of anaesthesia: many studies13,14,15,16 have concluded that the IVCCI is an independent predictor of hypotension after the induction of GA or spinal anaesthesia (SA), whereas other studies17,18 have reported that an increase in the IVCCI does not predict hypotension during SA.

In the absence of relevant research, the present study aimed to evaluate the predictive effect of perioperative IVCCI on PIH under GA combined with LSPB in elderly patients undergoing hip arthroplasty. Changes in the diameter of the common iliac vein (CIV) were also observed to determine the sympathectomy of the unilateral lower extremity after LSPB.

Methods

Sample size and study patients

Ethical approval for this prospective, observational study [Ethical Committee Number 2023-043-(1)] was provided by the Shanghai Jiaotong University Affiliated Sixth People’s Hospital on 06 May 2023. The study was conducted in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all participants. The trial was registered at Chinese Clinical Trials (registration number ChiCTR2300074603), and the first registration date was 10/08/2023. The sample size for this study was determined using PASS 11 software through a power analysis. The parameters set for the analysis of a single ROC curve with a one-sided hypothesis test included a null hypothesis area under the curve (AUC0) of 0.5, an alternative hypothesis area under the curve (AUC1) of 0.8419, an alpha level of 0.025, and a desired power of 0.90, with a sample allocation ratio of 1:1. This analysis indicated a minimum required sample size of 24 participants. Ultimately, we exceeded this target, enrolling a total of 40 cases, comprising 18 individuals in the positive group (hypotensive group) and 22 individuals in the negative group (non-hypotensive group).

The first patient was enrolled on 10 May 2023, and written informed consent was obtained from all eligible patients. The patients included in this study were aged at least 65 years, had an American Society of Anaesthesiologists (ASA) physical status of 1 to 3, fasted for more than 8 h before surgery, and were scheduled for hip arthroplasty. The exclusion criteria were refusal to participate, ASA physical status > 3, dyspnoea, decompensated heart failure, severe arrhythmia, significant valvular disease, block site infection, nerve dysfunction of the affected limb, mental disorders, inability to communicate, and diseases that cause increased intrathoracic or intraabdominal pressure disturbing IVC measurements (pneumothorax, pulmonary embolism, pleural effusion, pectus excavatum chest deformities, intestinal obstruction, acute peritonitis).

Trial design

Patients were screened and enrolled in the preoperative suite on the day of their scheduled procedures. Baseline blood pressure (BP), heart rate (HR) and oxygen saturation (SpO2) were recorded. Subsequently, peripheral venous access for fluid infusion was established. After the measurements of the baseline IVC and CIV diameter were performed, ultrasound-guided LSPB was performed on all patients. A successful blockade was confirmed by a pinprick test, and the measurements of the IVC and CIV diameters were repeated 15 min post-LSPB. Finally, the patients entered the operating room for GA induction.

The measurements of IVC and CIV, and the calculation of IVCCI

All patients were placed in the supine position, and placid breathing was required during IVC measurements. To visualize the IVC, an ultrasound system with a low-frequency convex array probe (Navis, Wisonic, China; 2–6 MHz) was used. A single anaesthesiologist who had undergone institutional training for ultrasound use under anaesthesia and had at least 5 years of experience in the field performed the scan. The measurements were made in M-mode in the subxiphoid transabdominal long-axis view. The maximum (IVCmax) and minimum (IVCmin) IVC diameters were identified approximately 1–2 cm below the entry point of the hepatic veins during a normal respiratory cycle at a rate of 12–15 breaths/min (Fig. 1a). The IVCCI was calculated as (IVCmax − IVCmin)/IVCmax and was expressed as a ratio. The CIV diameter was measured at the junction of the external and internal iliac veins in the long axis view (Fig. 1b). The IVCCI and CIV were measured twice: at baseline and 15 min post-LSPB.

Fig. 1
figure 1

Typical ultrasound images. (a) Ultrasonography of IVC diameter measurement; (b) ultrasonography of CIV diameter measurement; (c) ultrasonography of lumbar plexus block; QL = quadratus lumborum; ESM = erector spinae muscle; PS = psoas major; VB = vertebral body; TP = transverse process; d: ultrasonography of sacral plexus block.

Full size image

Definition of primary and secondary outcomes

This study’s primary outcome was to evaluate the ability of the IVCCI to predict PIH via ROC analysis. PIH was defined as a systolic blood pressure (SBP) below 90 mmHg or a ≥ 30% drop from the baseline. This categorized patients into hypotensive and non-hypotensive groups. During anesthesia, SBP measurements were taken at baseline,15 min post-LSPB, and 3 min post-induction. The secondary outcome was to observe changes in the IVC, CIV, and IVCCI dimensions. Measurements for CIV and IVC were taken at baseline and 15 min post-LSPB.

The ultrasound-guided LSPB procedures

All patients received a loading dose of 0.6 mg/kg dexmedetomidine intravenously for 10 min before ultrasound-guided LSPB was performed in the same lateral decubitus position. For the lumbar plexus block, a low-frequency ultrasound probe was placed above the iliac crest at the posterior axillary line to conduct a transverse scan, and the shamrock sign was observed20 (Fig. 1c). Using an “in-plane” technique, a puncture needle (22-gauge, 100 mm, Tuoren™, Henan, China) was inserted into the interfascial plane between the transverse process and psoas major muscles, and 30 ml of 0.33% ropivacaine was injected into the lumbar plexus nerves. For the sacral plexus block, we used a modified technique described by Taha21. The low-frequency ultrasound probe was placed transversely at the level of the posterior superior iliac spine, where the ala of the ilium could be demonstrated as a continuous hyperechoic line. The probe was slowly moved until a gap representing the greater sciatic foramen started to emerge. At this point, the sacral plexus appeared as a hyperechoic structure located medial to the posterior border of the ischium (Fig. 1d) and deep to the piriformis. The needle was inserted lateral to the probe via an “in-plane” approach, and 20 ml of 0.33% ropivacaine was injected.

For patients undergoing thrombus prevention or anticoagulation treatment, the guidelines state that short-term low-molecular-weight heparin should be used before surgery to reduce the risk of haematoma after LSPB22. If the patient experiences local anaesthetic poisoning, mild poisoning symptoms should be treated with oxygen inhalation and electrocardiogram monitoring until the symptoms disappear. When central and cardiac toxicity reactions occur, immediate tracheal intubation with pure oxygen ventilation should be performed, benzodiazepine drugs should be given for anticonvulsant treatment, and 20% fat emulsion at a loading dose of 1.5 ml/kg, followed by a maintenance dose of 0.25 ml/kg/min, should be administered. If the patient experiences insufficient blockade, failed block, or intrathecal block in the course of evaluation of the nerve block effect, the GA plan should be adjusted as appropriate and recorded as a change in anaesthesiological management, and these patients were excluded from statistical analysis.

The GA procedures

After the onset of LSPB (approximately 30 min after blockade), GA was performed via the use of 0.2–0.3 μg/kg sufentanil, 0.15 mg/kg cisatracurium, and 0.2–0.3 mg/kg etomidate, which was maintained via inhalation of sevoflurane (1–2 vol%) in oxygen-enriched air through endotracheal intubation. Patients who experienced prolonged airway instrumentation because of difficult intubation were excluded from further data analysis. SBP, diastolic blood pressure (DBP), MAP, and HR were measured at baseline, 15 min post-LSPB, and 3 min postinduction. When hypotension lasted for more than 2 min despite rapid fluid administration, an intravenous bolus of 5 mg ephedrine was given.

Statistical analysis

For the statistical analysis, we used SPSS version 23.0 for Windows (Armonk, NY: IBM Corp). The Shapiro‒Wilk test was used to test the normality of continuous variables. Continuous variables that were normally distributed are presented as the means ± standard deviations. Student’s two-sample t test was used for comparisons. Categorical variables are shown as percentages and absolute numbers of cases. The χ2 test and Fisher’s exact test were used for contingency table analysis as appropriate. Two-sided p values are shown, and the limit of statistical significance was set to p < 0.05. Receiver operating characteristic (ROC) curve analysis was used to investigate the associations of the IVCCI with hypotension after the induction of GA. We a priori decided that the IVCCI was clinically relevant if the area under the curve (AUC) was > 0.7.

Results

Population demographics and characteristics

This research involved 58 individuals. Six individuals met the exclusion criteria, and 12 patients were excluded from the analysis. Forty individuals were ultimately included in the trial, as shown in Fig. S1. In this study, there were no cases of haematoma, local anaesthetic poisoning or intrathecal block. This resulted in 40 patients being included in the statistical analysis. Among these 40 patients, depression of BP was common in almost all patients, but more notably, 18 patients (45%) developed PIH during GA. The demographics of the study population and a comparison of the participants by the presence or absence of PIH are provided in Table 1. All of the variables were similar between the two groups without any significant intergroup differences.

Table 1 Baseline population characteristics.
Full size table

ROC curve analysis (primary outcome)

No statistically significant differences in baseline nor post-LSPB IVCCI were detected comparing between hypotensive and non-hypotensive patients, shown in Table 2. The receiver operating characteristic (ROC) curve is presented in Fig. 2. The AUC of the ROC curve analysis of IVCCI for predicting hypotension was 0.54 (95% confidence interval: 0.35–0.72, P = 0.69). IVCCI does not predict PIH after induction of GA combined with LSPB in elder patients.

Table 2 Comparison of ultrasonography indices between the groups.
Full size table
Fig. 2
figure 2

ROC curve analysis of the IVCCI as a predictor of PIH. The AUC was 0.54.

Full size image

Changes in CIV and IVC after LSPB (secondary outcomes)

A significant expansion of the CIV (0.83 cm to 1.10 cm) was observed 15 min post-LSPB compared with baseline in all patients (p < 0.0001). No significant differences in the IVC diameter or IVCCI were detected between baseline and 15 min post-LSPB. The variations in the ultrasonography indices are shown in Table 2 and Fig. 3.

Fig. 3
figure 3

Comparison of ultrasonography indices between baseline and after block.

Full size image

Hemodynamic data

As shown in Fig. 4, all patients experienced a significant decrease in SBP, DBP, MAP, HR after induction of GA compared with parameters at baseline: the mean decrease in SBP was 39.10 ± 17.57 mmHg(26.26%), in DBP was 16.13 ± 10.53 mmHg(20.61%), in MAP was 23.80 ± 11.14 mmHg(23.37%), in HR was12.48 ± 5.37bmp(16.18%) (P < 0.0001). Meanwhile, no difference was detectable in hemodynamic statistics of 15 min post-LSPB compared with baseline.

Fig. 4
figure 4

Haemodynamic parameters of patients. (95% confidence interval [CI]: 0.35–0.72, P = 0.69).

Full size image

Discussion

This observational study revealed that the incidence of severe hypotension induced by GA combined with LSPB was 45% in elderly patients, which could not be predicted by the IVCCI. Additionally, significant expansion of the CIV was observed 15 min after LSPB, indicating sympathetic blockade of the unilateral lower extremity, but the decrease in the MAP from baseline was not statistically significant at the specified time.

Age greater than 50 years has been described as an independent predictor of hypotension during anaesthesia23,24. Not surprisingly, the incidence rate of PIH in the geriatric population of the current study was 45%, whereas it was 18.1% in a recent retrospective study of 2037 surgical patients aged 46 to 71 years25. The characteristic reduction in cardiovascular function with ageing explains the susceptibility to hypotension in elderly patients. First, the CO decreases by nearly 50% between the ages of 20 and 80 because of the loss of cardiac myocytes and interstitial fibrosis26. Second, vascular compliance decreases with age, leading to dysfunction of blood pressure regulation27. Additionally, preoperative hypovolemia caused by fasting, anaemia, or dehydration and continuation of ACE-I/ARBs28 also makes geriatric patients prone to PIH.

Zhang et al.13 described the predictive value of preoperative ultrasound IVCCI measurements in patients with hypotension after the induction of GA with a cut-off value of 43%. Salama et al.15 also obtained a similar cut-off value of 44.7% for predicting hypotension after the induction of SA. Nevertheless, Ceruti et al.16 and Mačiulienė et al.18 failed to detect a prognostic role of the IVCCI after the induction of SA in spontaneously breathing patients. In our study, the IVCCI was not effective in predicting hypotension after the induction of GA combined with LSPB in elderly patients. Several factors should be considered to explain our findings. First, and most importantly, the population of our study included elderly patients aged at least 65 years, who are more likely to develop PIH, as discussed above, despite a sufficient volume assessed by the IVCCI preoperatively. Second, a decrease in venous compliance decreases the variation in the IVC caused by inspiration and expiration29, leading to the IVCCI being unable to truly reflect the volume status in elderly patients. Third, the volume deficiency may have improved since the first IVCCI measurement due to the amount of fluid given before GA.

Recently, novel uses of ultrasonography in the internal jugular vein, subclavian vein and carotid artery were reported for the assessment of PIH in addition to the IVCCI. Among these methods, the most studied is the carotid artery corrected flow time (CFTc), which is the duration (in ms) of systole in the cardiac cycle adjusted for heart rate and remains unaffected by respiration30. In general, a reduction in the CFTc indicates better fluid responsiveness. However, the accuracy of CFTc in predicting volume status could theoretically be skewed by conditions frequently encountered in elderly individuals, such as reduced myocardial contractility, carotid stenoses and unanticipated carotid calcifications31,32. In conclusion, the predictive efficacy of the CFTc for PIH in geriatric patients requires further evaluation.

The updated evidence-based clinical practice guidelines recommend regional anaesthesia for the management of hip fractures in elderly patients33. Technical difficulty in the placement of a spinal needle due to the ossification of ligaments and bony bridges limits the application of spinal or epidural anaesthesia in elderly patients, and complications such as hypotension, urinary retention, and spinal haematoma should be considered34. Recently, different regional analgesic techniques, such as supra-inguinal fascia iliaca block (S-FICB), lumbar erector spinae block (L-ESPB) and pericapsular nerve group (PENG) block, have increased in popularity for hip surgery. Although the efficacy of these procedures has been demonstrated in pain management for hip surgery, their limitations have also been described34,35,36,37. A high volume (40 ml) of local anaesthetic (LA) is required for blockage of the lumbar plexus and its branches during S-FICB or L-ESPB, which could lead to the complication of local anaesthetic toxicity35,36. Moreover, the efficacy of S-FICB in blocking the obturator nerve has been investigated37, and no blockade of the lateral femoral cutaneous nerve was found after PENG block, which was required for the relief of incisional surgical pain38. In summary, preventing the potential cardiac toxicity of high-volume LA and the haemodynamic disturbances caused by inadequate analgesia perioperatively is essential, especially for geriatric patients with diminished cardiovascular function. Consequently, an LSPB was selected for pain management during hip arthroplasty in the present study.

Vasodilation in the forearm arteries has been reported after brachial plexus block39,40. However, to the best of our knowledge, there is a lack of systematic data on sympathetic blockade of the lower extremity after lumbar or sacral plexus block. The current study provided limited but valuable relevant data: a significant expansion of CIV (0.83 cm to 1.10 cm) was observed 15 min post-LSPB. Owing to the sympathetic innervation of the lower extremities41, we assumed that the vasodilation of CIV was the consequence of blockade of the postganglionic fibre branches of the femoral and genitofemoral nerves under the lumbar plexus block.

In the present study, the expansion of the unilateral CIV caused by LSPB did not result in a significant decrease in the MAP. Compared with SA-induced vasodilatation caused by blockade of preganglionic fibres in both lower extremities41, the level of sympathetic denervation induced by LSPB is considered to be much lower than that induced by SA because of the unilateral postganglionic fibre blockade. In addition to the limited preload reduction resulting from vasodilation of CIV, the high degree of resting sympathetic tone in elderly patients42 may explain the stabilization of blood pressure, as arterial compensatory contraction resists relative intravascular volume reduction. However, Visme et al.9 reported that in elderly patients ranging in age from 68 to 97 years, the initial decrease in MAP was 38% in the SA group and 27% in the LSPB group, not significantly different. We believe that the significant drop in blood pressure in the previous study may be related to the consumption of alfentanil following LSPB, which was not administered in our research.

Previous cases of lumbar plexus block causing intrathecal block have been reported. Stevens et al.4 reported the occurrence of epidural block after lumbar plexus block, with symptoms of bilateral block. Bradley et al.43 presented a case of subdural spread of local anaesthetic during a lumbar plexus block, resulting in hypotension, incomplete motor block, and sensory block peaking at the C7 level. These reported complications were not observed in this trial and may be related to the small sample size and ultrasound-guided nerve block; however, attention should be given to these possible complications. Anatomically, PENG block is a nonsympathetic block that specifically targets the distal sensory branches of the anterior portion of the joint capsule38, and the incidence of epidural block could be prevented because the target of PENG blockade is at a considerable distance from the intervertebral foramina. Further studies with larger sample sizes are needed to test these hypotheses.

The present study has several limitations. First, this was a single-centre study with a relatively small sample size. These results must be considered preliminary and need to be validated in future studies. Second, we did not measure the IVCCI after the induction of GA, as spontaneous respiration had been replaced by positive pressure ventilation. Third, the elderly patient population was treated with heterogeneous types of antihypertensives, such as calcium channel blockers (CCBs), ACEIs/ARBs or beta blockers, before surgery. We did not compare the effects of different antihypertensive drugs on PIH. Fourth, fasting hours greater than 8 h in elderly patients were associated with dehydration and a potentially increased incidence of PIH in the current study. Finally, the IVC moves in both the mediolateral and craniocaudal directions during respirophasic ultrasound imaging44. More pain resulting from hip fracture than from hip arthritis can induce rapid-shallow breathing, which could cause minor changes in IVC variations. In the present study, placid breathing was provided to the patient during IVC ultrasound imaging to reduce measurement bias.

Conclusion

The IVCCI is not an effective predictor of severe hypotension after the induction of GA combined with LSPB in elderly patients undergoing hip arthroplasty. Additionally, significant expansion of the CIV was observed 15 min after LSPB, indicating sympathetic blockade of the unilateral lower extremity. Although the high degree of resting sympathetic tone in elderly patients keeps their blood pressure stable, vasodilator effects produced by LSPB may increase the incidence and severity of PIH. Future studies must evaluate whether LSPB is an independent risk factor for hypotension after the induction of general anaesthesia in elderly patients.