Peridural Anesthesia or Ultrasound-Guided Continuous 3-in-1 Block

Abstract

Purpose:

The aim of our study was to investigate the clinical relevance of preoperative acute pain management and cardiovascular stability by ultrasound-guided continuous 3-in-1 nerve block in very elderly patients with hip fracture when compared to epidural anesthesia (PDA).

Methods:

To study the analgesic effect, we enrolled 37 very elderly patients with hip fractures, of whom 3 patients with dementia had to be excluded. Thus, 34 patients were randomized to 1 of the 3 groups: group A (ultrasound-guided continuous 3-in-1 block, bupivacaine; n = 10, dropout rate: 0), group B (PDA, bupivacaine; n = 14, dropout rate: 8), and group C (systemic pain therapy, piritramide/paracetamol; n = 10, dropout rate: 0). Pain intensity was assessed preoperatively and up to 24 hours postoperatively using a visual analog scale, verbal rating scale, analgesic consumption, scale of well-being, and cardiocirculatory parameters (eg, serum troponin T).

Results:

Our data show that in the preoperative period both regional anesthesia (RA) procedures (analgesia responders after 1 hour: 86.7% and 100%; P = .001) were superior to systemic analgesia (analgesia responders: 46.7%), and the rescue medication requirement in the 2 RA groups was significantly lower (P = .02). Serum troponin T level increased only in the systemic analgesia group (P = .04). In the emergency department, the disadvantage of PDA in geriatric patients with hip fracture was the fact that procedures were more complex, resulting in a high dropout rate (57.1%). The use of PDA has to be critically discussed for ethical concerns.

Conclusion:

In the specific situation of acute hospital admission, the ultrasound-guided continuous 3-in-1 block appears to be indicated as a stress-free means of providing adequate preoperative pain relief in very elderly patients with hip fracture. However, these findings should be corroborated by studies involving larger numbers of patients.

Background

Hip fracture is one of the most common injuries in the very elderly individuals and is treated best by comanagement involving trauma surgeons, anesthesiologists, and geriatric specialists.^1,2 In the acute phase of hospitalization, all medical problems of very elderly patients with hip fracture, such as health status, comorbidities, anesthesia, surgical procedure, timing, and organization, should be addressed in time, because long-lasting immobilization can lead to serious problems such as pneumonia, disuse atrophy, and deep vein thrombosis.² However, even with rapid surgical treatment, high morbidity and mortality are expected due to the advanced age and comorbidities of these patients.³ Therefore, it is recommended that surgery be scheduled within 24 hours^2,4 and an adequate choice of anesthesia be addressed.⁵ Meanwhile, adequate pain relief after hip arthroplasty should be achieved with either systemic opioids or regional anesthesia (RA) methods, namely epidural anesthesia (PDA) or continuous femoral nerve sheath block.⁶ The continuous 3-in-1 block is a procedure that is very easy to handle, has fewer side effects, reduces pain scores and analgesic requirements in geriatric patients with hip fracture, ^7–9 and has been critically reviewed.¹⁰ Nowadays, the quality of the sensory block in all the 3 nerves is significantly improved by using the ultrasound-guided technique instead of the nerve stimulator-assisted method.^11–13 In contrast, PDA is a technically more complex procedure with good analgesic potential, but more potential complications such as local anesthetic neurotoxicity, hypotension, bradycardia, accidental dural puncture, and transient paresthesis.^14,15 It is noteworthy that Matot et al found in patients with hip fracture who have or are at risk of coronary artery disease, a favorable influence of PDA on perioperative cardiac situation under PDA.¹⁶ Therefore, we need to ask whether a PDA or a 3-in-1 block is indicated for analgesia in these patients. It was thus of interest to evaluate very old patients with hip fracture concerning preoperative analgesia and cardiovascular stability provided by 3-in-1 block, PDA, or systemic analgesia.

The aim of this study was to examine preoperative analgesia in patients with geriatric hip fracture immediately after arrival in the emergency department, as induced by continuous 3-in-1-block, PDA, and systemic opioids (control group).

Study Design and Methods

Patients

After the study was approved by the ethics committee of Innsbruck Medical University all patients gave their written informed consent. This prospective, single-center, controlled, double-blind, randomized pilot study was performed in geriatric ASA I-III patients (American Society of Anesthesiologists) patients scheduled for acute hip fracture surgery. Inclusion criteria were very elderly patients with age >80 years who gave informed consent for spinal anesthesia, PDA, and continuous 3-in-1 block. The following exclusion criteria were relevant: known intolerance or allergies to drugs, planned or required general anesthesia, refusal of consent, participation in a different study, administration of midazolam as premedication, patient with chronic pain, contraindications and spinal anesthesia failure, or incomplete data records. A score of <18 on the Mini-Mental State Examination (MMSE) was considered an exclusion criterion.¹⁷ Furthermore, those patients whose surgery did not take place within 36 hours or for whom the time needed to conduct the study would have meant rescheduling the surgery were excluded.

Study Design and Group Assignment

On admission to the hospital and after diagnosis including indication for surgical treatment the anesthesia team conducting the study was called in. After preoperative anesthesiological examination (cardiopulmonary condition, comorbidities) and assessment of exclusion criteria in each patient, the patient was informed of potential medical complications (such as embolism, hemorrhage, allergy, high spinal anesthesia, spinal damage, etc) and side effects in connection with general anesthesia and RA. All patients gave their written informed consent. The patients were randomized according to a computer-generated randomization list. After installing monitoring equipment (electrocardiogram [ECG], oscillometric blood pressure cuff [RR], pulse oxymetry), intravenous (iv) infusion with 500 mL Ringer lactate was commenced and the patient received pain management with an initial dose of 0.05 mg/kg piritramide administered iv. Regional anesthesia was induced by the anesthesia team not involved in data collection.

As decided by randomization, patients underwent pain therapy in 3 groups:

• Group A: for preoperative analgesia, patients received a continuous 3-in-1 block (femoral, lateral cutaneous, obturatorius) using the ultrasonographic guidance technique (SonoSite Micromaxx, SoMB Technology, linear array transducer 13-6 MHz, Sittendorf, Austria).^12,13,18 The catheter was inserted at least 12 to 15 cm into the femoral sheath and its position was sonographically estimated in order to ensure a higher success rate for the obturatorius block. The initial bolus administered was 30 mL 0.25% bupivacaine (Bucain, Actavis Group PTC, Island) followed by 0.125% bupivacaine 6 mL/h per perfusion. Quality of sensory block was assessed using the pinprick test to establish that the block reached all 3 main nerves (femoral, lateral cutaneous, obturatorius). The motor block was not tested because we wanted to avoid any dislocation of the patient’s femoral fracture.¹² If no response was achieved, an additional 10 mL bolus of 0.125% bupivacaine was administered. For surgery, spinal anesthesia was decided on and was induced in a sitting position. However, low-dose ketamine S(+) isomer (Ketanest-S, Pfizer Corporation, Vienna, Austria) 0.1 to 0.2 mg/kg was administered for pain relief when needed. After subcutaneous (sc) infiltration with 2 mL lidocaine 2% using a median or lateral approach, a 27G pencil-point needle (Braun, Melsungen, Germany) was introduced at the L3/L4 or L4/L5 interspace until free backflow of cerebrospinal fluid was observed. Patients received isobaric bupivacaine 0.5% in a total volume of 2.5 to 4.0 mL without any adjuncts injected into the subarachnoid space over 1 minute. After 10 minutes, patients were returned to the supine position. Perioperatively, ephedrine, midazolam, or both were administered iv, if needed.

• Group B: for preoperative analgesia, patients in a sitting position underwent PDA using a Tuohy needle at L3/L4 or L4/L5 with the “loss of resistance” technique to locate the epidural space. A test bolus of 2 mL 0.25% bupivacaine, followed by 0.25% bupivacaine, was administered in 2-mL steps to achieve the appropriate level of analgesia at Th12 (approximately 10-12 mL total). Subsequently, patients received 0.125% bupivacaine 8 mL/h per perfusion. To achieve the necessary anesthesia level a single 4 mL bolus of 0.125% bupivacaine could be additionally administered. Sensory level was tested using the pinprick test. Perioperatively, ephedrine, midazolam, or both were administered iv, if needed.

• Group C (control): for preoperative analgesia, patients received an initial dose of 0.05 mg/kg piritramide iv and additionally systemic analgesics, either piritramide 3 mg sc as a bolus, or paracetamol (Perfalgan, Bristol-Myers Squibb, Vienna, Austria) 1 g as a short infusion with a maximal daily dose of 3 g, if needed. For surgery, spinal anesthesia was performed as described above.

Patients were subsequently observed preoperatively and monitored by ECG, RR, and pulse oxymetry in the nearby recovery room for at least 2 hours under the supervision of the anesthesiologist. Thereafter, depending on surgery scheduling the patients were transferred to the operating room. For surgery, spinal anesthesia was induced in Groups A and C, while in group B PDA was achieved with bupivacaine 0.5% up to an analgesic level of TH 10.

As first-line rescue medication, patients received either paracetamol 1 g as a short infusion or piritramide (Dipidolor, Janssen-Cilag Pharma GmbH, Vienna, Austria) 3 mg iv, if required. Administration procedures were defined by hourly monitoring of visual analog scale (VAS) as follows: for VAS >3 piritramide 3 mg sc or paracetamol 1 g was administered in a short infusion. If it was not successful within 30 minutes (VAS <3), pain relief was supplemented with additional doses of bupivacaine administered via catheter or repeated doses of piritramide iv in agreement with the local pain service. In addition, it should be noted that piritramide is a long-lasting, strong opioid with analgesic effects similar to those of morphine but not more side effects than morphine, for example, respiratory depression, vomiting, and nausea^19,20 It is especially designed for prehospital and inhospital pain management.^20,21

Variables

Demographic data

Patient age, gender, body mass index (BMI), MMSE, time to surgery (TTS), duration of surgery (DOS), and length of stay (LOS) were recorded.

Pain scores

For pain assessment in geriatric patients easy-to-understand numerical pain scores (NPS) and VAS were used, both at rest and during passive movement of the leg. The VAS was a horizontal line of 100 mm in length, running from left to right and labeled “no pain” (0 mm) to “incredibly intense pain” (100 mm). The baseline assessments were performed in the emergency department before inducing RA or systemic pain control. Thereafter, the pain score at rest and in motion were measured 1 and 12 hours after administration of RA, immediately preoperatively, as well as 3 hours and 24 hours postoperatively. An analgesia responder was defined as a patient with a reduction of >50% on the pain score within 1 hour. Additionally, a mood rating was performed using a verbal mood scale from 1 (very bad) to 5 (excellent). Assessment was performed in the order specified, namely VAS, mood, and NPS.

Additional recorded data

Rescue pain medication and side effects were recorded. In addition, serum samples of troponin T, creatine kinase (CK), CK isoenzyme cardiac fraction MG (CK-MB), lactate, and cortisol (as stress parameters) were calculated.

Statistical Analysis

Primary outcome was the analgesic effect and secondary outcome was the cardiovascular stability. Null hypothesis H0 was defined as no difference between the groups. For sample size estimation, previous publications^6,16 were used based on the following assumptions: the level of significance (error probability) was set at 0.05; the power of the test at 80%; and the difference between the groups was defined as 2 cm on the 10-cm VAS scale with a standard deviation of ±2 cm. Thus, the sample size was estimated as 22 patients/group including a dropout rate of 10%.

Interim analysis was scheduled when 50% of the patients had been studied. In the geriatric patients with acute trauma in group B (PDA) psychological, medical, and technical problems were encountered during PDA catheter placement in the emergency department, resulting in a high dropout rate. Consequently, after the scheduled interim analysis the study was discontinued.

Analyzed data were then checked for normality. Nonparametric data were analyzed descriptively, and the Mann-Whitney U Test and the Kruskall-Wallis test were performed to detect significant differences among groups, as appropriate. A P value of .05 was deemed statistically significant throughout the study.

Results

Demographic Data

A total of 37 patients were ultimately included in the study. Three patients had an MMSE of 10 and were excluded. In group A (3-in-1 block) and group C (control) each, 10 patients completed the study program. In group B (PDA), 14 patients were randomized, of whom 8 (57.1%) were dropped from the study in the emergency department because of no-go criteria. Three stressed patients ultimately rejected PDA without reason after randomization and allocation. In all, 1 patient revealed a contraindication to PDA (anticoagulation therapy) after randomization and in 4 very elderly patients PDA had to be terminated because of unknown pathological vertebral column changes that caused difficulties involving catheter placement and subjective stress reactions. It should be noted that some hours later in 2 of the latter patients, spinal anesthesia was easily performed in the operating theater for surgical treatment of the femoral neck fracture, while the other patients underwent general anesthesia. Thus, this study was discontinued for no-go reasons solely in group B since PDA in the acute phase of hospitalization was not clinically acceptable in very elderly patients (Figure 1).

Figure 1.

Study design.

After discontinuation of the study, interim analysis showed the groups to be comparable regarding age, height, BMI, TTS, LOS, DOS, and MMSE. It was striking that the study patients included only 3 men (1 in group A and 2 in group B, Table 1).

Table 1.

Demographic Data: Age; BMI, Body Mass Index; MMSE, TTS, DOA, DOS.

	Group A	Group B	Group C
Age	87.0 ± 10.1	82.2 ± 8.3	91.1 ± 9.0
ASA	2.8 ± 0.4	2.7 ± 0.5	2.7 ± 0.5
BMI	22.6 ± 5.2	20.3 ± 2.6	24.9 ± 1.7
MMSE	28.1 ± 3.0	28.0 ± 1.4	28.0 ± 1.7
Diagnose
Pertrochanteric femur fracture	6 (60%)	5 (83.3%)	3 (30%)
Medial femur neck fracture	4 (40%)	1 (16.7%)	7 (70%)
Surgery
Hemiarthroplasty	3 (30%)	0	4 (40%)
Total hip replacement	0	1 (16.7%)	2 (20%)
Dynamic hip screw	3 (30%)	1 (16.7%)	3 (30%)
Cannulated screws	1 (10%)	0	0
Proximal femoral nail	3 (30%)	4 (66.6%)	1 (10%)
TTS	26.8 ± 18.6	18.5 ± 11.4	24.9 ± 1.7
DOS	75.7 ± 26.2	73.3 ± 20.8	76.4 ± 27.3

Abbreviations: BMI, body mass index; MMSE, Mini-Mental State Examination; TTS, time to surgery; DOA, duration of anesthesia; DOS, duration of surgery; ASA, American Society of Anesthesiologists.

Pain Measurement

When measuring pain at rest and in motion those patients who received RA (groups A and B) showed a significantly better analgesic effect; moreover, mood was improved in all the groups. As shown in Figure 2 for VAS and verbal pain scale (VPS), baseline pain intensity at rest and in motion was comparable. Already 1 hour after RA significant differences were seen between the treatment groups (analgesia responders at rest and in motion: group A: 86.7% ± 5.8% and 86.7% ± 5.8%; group B: 100% and 100%; control; group C: 70% ± 17.3% and 46.7% ± 15.3%, respectively; Figure 2). This significant difference was observed throughout the preoperative phase. No significant differences were seen in the course of postoperative pain. On the mood scale, a significant improvement in treatment was seen in the groups only 24 hours after the surgery (Figure 2). Measuring VPS in motion showed an analgesic advantage for the RA groups; at rest, this effect was seen only as a trend (Figure 2).

Figure 2.

Patients with geriatric hip fracture and acute preoperative pain management. Pain measured at rest and in motion using the visual analogue scale (VAS) and the verbal pain scale (VPS): group A: 3-in-1 block, group B: epidural anesthesia (PDA), and group C: systemic analgesia. Mood measurement (5: excellent, 1: very bad). Mann-Whitney U test: P < .05; between-group comparison: ^XGroup A versus group C; *group B versus group C; °group A versus group B.

As seen from Table 2, consumption of additional analgesics was elevated in the control group when compared to the treatment groups, an effect that is significant for paracetamol and only a trend for piritramide. It should be noted that the need for systemic analgesics was lowest in group A. In the postoperative period, a significantly increased need for paracetamol was seen in the treatment groups (P = .02), but not for piritramide (Table 2).

Table 2.

Need for Additional Systemic Analgesics: Piritramide (mg/d) and Paracetamol (g/d), as well as Serum Levels of Troponin T (µg/L), CK, Serum Creatine Kinase Isoenzyme Cardiac Fraction MG (CK-MB), Cortisol (µg/L), and Lactate (mg/dL) in Patients With Hip Fracture in the ER, Preoperative, and Postoperative Phase: 3-in-1 Block (Group A), PDA (Group B), and Systemic Analgesia (Group C).^a

Group	ER	Preoperative	Postoperative
Paracetamol
A	0.2 ± 0.4	0.1 ± 0.32b	0.3 ± 0.7c
B	0.2 ± 0.5	0.0 ± 0.0b	0.8 ± 1.8c
C	0.1 ± 0.32	1.7 ± 1.4	1.1 ± 1.3
Piritramide
A	0	0.75 ± 2.4	0.75 ± 2.4
B	0	1.3 ± 3.1	0.5 ± 1.2
C	0	3.4 ± 5.1	3.0 ± 5.2
Troponin T (<0.1 mg/L)
A	0.01 ± 0	0.01 ± 0b	0.025 ± 0.04
B	0.01 ± 0	0.01 ± 0b	0.01 ± 0b
C	0.011 ± 0.005	0.018 ± 0.01	0.024 ± 0.02
Serum CK (<174 U/L)
A	228.2 ± 222.2	249.2 ± 208.5	354.3 ± 125.8
B	220.6 ± 199.8	252.2 ± 209.5	361.8 ± 205.4
C	118.2 ± 82.4	142.3 ± 117.7	349.9 ± 178.8
Serum CK-MB (0-24 U/L)
A	16.0 ± 5.9	15.7 ± 3.9	23.4 ± 12.2
B	21.0 ± 13.5	21.3 ± 13.1	19.5 ± 7.8
C	18.1 ± 11.4	17.6 ± 6.9	25.2 ± 17.7
Cortisol (am 140-690 nmol/L; pm 80-330 nmol/L)
A	192.4 ± 71.9	191.1 ± 67.6	250.3 ± 81.7
B	285.5 ± 72.8	234.7 ± 15.6	219.3 ± 24.8
C	220.0 ± 134.0	211.6 ± 115.1	260.7 ± 120.3
Serum lactate (5.7-22 mg/dL)
A	14.1 ± 4.2	11.8 ± 7.2	13.2 ± 6.0
B	15.2 ± 4.0	14.8 ± 8.1	10.9 ± 7.0
C	15.0 ± 5.7	16.5 ± 8.3	15.1 ± 8.9

Abbreviation: RA, regional anesthesia; CK, creatine kinase; CK-MB, creatine kinase isoenzyme cardiac fraction MG; ER, emergency room; SD, standard deviation.

^a Normal values listed in parenthesis as appropriate. Mean ± SD; Mann-Whitney U test: P < .05.

^b Between-group comparison, group A or B versus group C.

^c Comparison between pre- and postoperative value.

Cardiovascular Stability

As shown in Table 2, significant differences in serum troponin T levels were seen between the treatment groups and the control groups (Table 2). In the pre- and postoperative phase, the maximum serum troponin T levels were significantly lower in patients in group A (P = .03) and group B (P = .02) when compared to placebo. Only in group C there was a nonsignificant increase in the troponin T levels recognized over time (P = .07). It is noteworthy that only 1 patient in group A showed a postoperative nonsignificant increase in troponin T to 0.13 mg/L. This increase was clinically without signs of cardiopulmonary disease and significantly declined within 24 hours. In group B, we did not observe any abnormal troponin T levels. In the control group, however, 5 patients with normal troponin T values at baseline (0.01 mg/L) showed a slight increase to 0.071 mg/L. In this group, the troponin T value at baseline already slightly increased (from .015 to .018 mg/L) in 2 patients, but subsequently showed a declining tendency (from 0.012 to 0.017 mg/L).

In serum levels of CK, CK-MB, as well as the stress indicators lactate and cortisol no significant differences between the treatment groups and the control group were observed (Table 2). However, only in group C a significant increase in CK was observed over time (P = .005), while CK-MB discretely remained in the normal range (P = .7). These CK and CK-MB findings lay the groundwork for suspecting that this effect is nonspecific and can be closely interpreted in connection with the preexisting trauma. Under simultaneous consideration of the troponin T values these CK and CK-MB readings can be viewed as insensitive parameters for answering the second hypothesis of possible heart involvement.

Discussion

In the emergency department, we found that for the preoperative acute pain management of very elderly patients with hip fracture (1) RA (continuous 3-in-1 block and PDA) appears to be superior to systemic pain therapy, as seen in the pain scores, and (2) an increase in troponin T levels was observed only in group C (systemic analgesia). Moreover, (3) the need for additional analgesics (rescue medication) was less strong in the RA groups, while a considerable significance could be detected only for paracetamol. It should be noted that (4) in patients with acute geriatric hip fracture the more complex PDA procedure and catheter placement resulted in technical procedure problems and several unsuccessful attempts. These difficulties ultimately led to the high dropout rate and premature termination of the study. Our experience, which raised many technical and ethical concerns, stands in contrast to the study by Matot et al, who favored PDA for geriatric patients with hip fractures; that study, however, was performed in patients with coronary artery disease aged over 65 years.¹⁶ Asao et al confirm our concerns in very elderly patients with hip fracture, stating that they prefer not to administer PDA because they experienced more difficulties and several failed attempts during the procedure.²² Moreover, an additional advantage of the ultrasound-guided 3-in-1 block is the fact that it is easy to perform in a lying position without entailing major technical problems, and afterward a pain-free sitting position can be taken to induce spinal anesthesia for surgery.²³

Nevertheless, these interim data are worth discussing. The advantage of RA over systemic analgesia in the emergency department was already observed in some studies.^9,24 Bruce Martin summarized 4 studies that compared the 3-in-1 block with systemic analgesia, indicating more effective pain reduction with fewer side effects and less consumption of systemic analgesics when using a 3-in-1 nerve block.⁸ However, many of these studies were performed with the single-shot technique.^8,9 Other studies concluded that a continuous 3-in-1 block catheter is the procedure of choice for acute pain management in patients with femoral neck fracture in the preoperative setting.^9,10,24,25 Additionally, we agree that this procedure (catheter technique) has the dual advantage of providing early analgesia and continued postoperative pain management,²⁶ while not interfering with anticoagulatory prophylactic therapy. However, some studies were performed with the nerve stimulation-assisted single-shot technique, while others used the ultrasound-guided technique.^12,13,19,25 Nevertheless, it should be kept in mind that the obturator nerve cannot be easily reached with the electrical neurostimulation technique and success rates were low.²⁷ Nevertheless, our 1-hour analgesia responder rate of >86.7% achieved with the ultrasound guidance technique was comparable to that of other studies.^12,18,22

Our very elderly patient population undergoing either ultrasound-guided continuous 3-in-1-block or PDA showed no changes in troponin T levels. Only those undergoing systemic analgesia exhibited increased troponin T levels without clinical signs. This observation of a reduced incidence of cardiac ischemia in the RA groups is also supported by others.^9,16,28 Our study was able to confirm that in geriatric patients with hip fracture the ultrasound-guided continuous 3-in-1 nerve block provided clinically sufficient preoperative pain therapy with cardiovascular stability as verified by laboratory chemistry methods.

As limitations we must stress that this observation was designed as a pilot study, but was discontinued when some patients showed no-go criteria, thus raising ethical concerns. Thus, although this study provides valuable data, our interim analysis is based on a small number of cases. For this reason it is difficult to reach definite conclusions. Moreover, being aware of the widespread bias issue involved in such small trials we employed a well-defined sequence of pain assessment. Nevertheless, in the clinical setting a minor bias can never be entirely ruled out, although in our study it appears to be very unlikely.²⁹ Needless to say, we took every precaution in interpreting the results of our pilot study, a fact emphasized by us by recommending further studies. Despite all limitations and with all due respect, the unfavorable and negative result of not using PDA in the elderly individuals on emergency room admission remains a topic of interest. Furthermore, the types of surgical procedure were not stratified but were numerous and all resulted in a wound area of similar size. Thus, it is unlikely that for this question the surgical approach may influence our results. However, in light of all these limitations it is strongly recommended that a follow-up study should be conducted with a larger number of patients and an appropriate control group, for example with sham catheter, but without a PDA group.

Conclusion

In the acute phase of hospitalization, there is some evidence that the use of the continuous 3-in-1 block in very elderly patients with hip fracture appears to be beneficial in providing sufficient preoperative pain management with cardiovascular stability. The use of PDA in these geriatric patients with hip fracture should be critically evaluated and, from our experience, it is not recommended in an acute emergency situation because of technical concerns. However, this pilot study has to be evaluated by further studies involving a larger number of patients and an appropriate control group.