SciELO - Scientific Electronic Library Online

vol.58 issue2In vitro and in vivo neuromuscular effects of atracurium and rocuronium in rats treated with carbamazepine for seven daysAccidental subarachnoid administration of 4 mg of morphine: case report author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Revista Brasileira de Anestesiologia

Print version ISSN 0034-7094On-line version ISSN 1806-907X

Rev. Bras. Anestesiol. vol.58 no.2 Campinas Mar./Apr. 2008 



Phrenic nerve block after interscalene brachial plexus block. Case report*


A bloqueo del nervio frénico después de la realización de bloqueo del plexo braquial por la vía interescalénica. Relato de caso



Luis Henrique CangianiI; Luis Augusto Edwards RezendeII; Armando Giancoli NetoIII

IAnestesiologista do Centro Médico de Campinas e do Instituto Penido Burnier
IIME3 (2006) do CET/SBA do Instituto Penido Burnier e Centro Médico de Campinas
IIIME3 (2007) do CET/SBA do Instituto Penido Burnier e Centro Médico de Campinas

Correspondence to




BACKGROUND AND OBJECTIVES: Phrenic nerve block is a common adverse event of brachial plexus block. However, in most cases it does not have any important clinical repercussion. The objective of this work was to report a case with phrenic nerve block with respiratory repercussions in a patient with chronic renal failure who had an extensive arteriovenous fistula created under perivascular interscalene brachial plexus block.
CASE REPORT: A 50-year old male patient, smoker, with chronic renal failure on hemodialysis, hypertension, hepatitis C, diabetes mellitus, and chronic obstructive pulmonary disease, was scheduled for creation of an arteriovenous fistula in the right upper limb under interscalene brachial plexus block. The brachial plexus was identified by a peripheral nerve stimulator. Thirty-five milliliter of a local anesthetic mixture containing equal parts of 2% lidocaine with epinephrine at 1:200.000 and 0.75% ropivacaine were injected. After the injection, the patient was alert and oriented, but developed dyspnea and predominance of intercostal respiration on the side of the blockade. Breath sounds were not present in the right base. SpO2 was maintained at 95% with oxygen through nasal cannula. Institution of invasive ventilatory support was not necessary. A chest X-ray showed the right hemidiaphragm on the 5th intecostal space. The patient returned to normal after three hours.
CONCLUSION: In this case, the patient developed complete paralysis of the phrenic nerve with respiratory symptoms. Although invasive treatment was not necessary, it is necessary to alert anesthesiologists to restrict the indication of this technique.

Key Words: ANESTHETIC TECHNIQUES, Regional: brachial plexus block; ANESTHETICS, Local: lidocaine, ropivacaine; COMPLICATIONS, Respiratory: phrenic nerve block; DISEASES, Renal: chronic renal failure; SURGERY, Vascular: arteriovenous fistula.


JUSTIFICATIVA Y OBJETIVOS: El bloqueo del nervio frénico es un evento adverso del bloqueo del plexo braquial, sin embargo, en su mayoría, sin repercusiones clínicas importantes. El objetivo de este relato fue presentar un caso en que ocurrió bloqueo del nervio frénico, con comprometimiento ventilatorio en paciente con insuficiencia renal crónica, sometido a la instalación de fístula arterio-venosa extensa, bajo bloqueo del plexo braquial por la vía perivascular interescalénica.
RELATO DEL CASO: Paciente del sexo masculino, 50 años, tabaquista, portador de insuficiencia renal crónica en régimen de hemodiálisis, hipertensión arterial, hepatitis C, diabetes melito, enfermedad pulmonar obstructiva crónica, sometido a la instalación de fístula arterio-venosa extensa en el miembro superior derecho bajo bloqueo de plexo braquial por la vía interescalénica. El plexo braquial fue localizado con la utilización del estimulador de nervio periférico. Se inyectaron 35 mL de una solución de anestésico local constituida de una mezcla de lidocaína a 2% con epinefrina a 1:200.000 y ropivacaína a 0,75% en partes iguales. Al final de la inyección el paciente estaba lúcido, pero sin embargo con disnea y predominio de incursión respiratoria intercostal ipsilateral al bloqueo. No había murmullo vesicular en la base del hemitórax derecho. La SpO2 se mantuvo en un 95%, con catéter nasal de oxígeno. No fue necesaria la instalación de métodos de auxilio ventilatorio invasivo. La radiografía del tórax reveló que el hemidiafragma derecho ocupaba el 5° espacio intercostal. El cuadro clínico se revirtió en tres horas.
CONCLUSIONES: El caso mostró que hubo parálisis total del nervio frénico con síntomas respiratorios. A pesar de no haber sido necesaria la terapéutica invasiva para el tratamiento, queda el aviso aquí para la restricción de la indicación de la técnica en esos casos.




Brachial plexus block is used in surgeries of the upper limbs, both in hospitalized and outpatients 1. It is a good alternative for severely ill patients, who benefit from the blockade instead of general anesthesia. However, a few adverse events have been described like pneumothorax, nerve lesions, accidental intravascular injection, Horner's syndrome and phrenic nerve block 2.

Some authors admit that phrenic block is always present when the perivascular interscalene technique is used but most of them do not have clinically important repercussions being probably secondary to partial nerve block 3. Other authors observed that varying degrees of phrenic nerve block also occur with the supraclavicular and infraclavicular techniques with few clinical manifestations 4,5.

The objective of this report was to present a case of phrenic nerve block with ventilatory compromise in a patient with chronic renal failure who had an extensive arteriovenous fistula created under interscalene brachial plexus block.



A 50-year old male patient, 1.77 m, 76 kg, physical status ASA III, was scheduled for the creation of an arteriovenous fistula in the right upper limb under perivascular interscalene brachial plexus block.

The patients was a smoker, had chronic renal failure on hemodialysis three times a week, hypertension, hepatitis C, diabetes mellitus and chronic obstructive pulmonary disease with predominance of the emphysematous pattern. He was alert and oriented, ruddy, hydrated, eupneic, blood pressure 180 x 100 mmHg, heart rate 75 bpm, normal sinus rhythm, normal heart sounds without murmurs, breath sounds present bilaterally but universally decreased without rales. Laboratory data revealed: hemoglobin 12.5 mg.dL-1, hematocrit 39.5%, platelets 105.000, Na+ 135 mEq.L-1, K+ 4.5 mEq.L-1, BUN 81 mg.dL-1, creatinine 3.4 mg.dL-1, glucose 203 mg.dL-1, prothrombin time 12.9 s, prothrombin activity 79%, activated partial thrombin time 28.4 s and INR 1.18.

Pre-anesthetic medication was not administered. In the operating room, venipuncture with a 20G catheter was performed in the left upper limb and normal saline (500 mL) was administered to keep the vein opened. Monitoring consisted of pulse oximetry, pletismograph, cardioscope on DII and V5 derivations and automatic non-invasive sphygmomanometer. The patient was sedated with intravenous fractionated doses of midazolam (5 mg) and fentanyl (50 µg). Oxygen (2 L.min-1) was administered via a nasal cannula. The patient was calm, cooperative, with good ventilation, with SpO2 of 96%. He was positioned for the blockade with his head rotated to the opposite side and right arm close to the body. The area was cleaned and sterile fields were placed around the area of the puncture.

The interscalene technique of Winnie was used to approach the brachial plexus and the site of the injection was determined by the peripheral nerve stimulator. Adequate stimulation caused a motor response below the shoulder with stimulus of up to 0.4 mA. The syringe was aspirated and the local anesthetic solution constituted of equal parts of 2% lidocaine with epinephrine at 1:200,000 and 0.75% ropivacaine was injected. A total of 35 mL of the anesthetic solution were injected.

Shortly after the administration of the local anesthetic was initiated, the stimulus evoked by the nerve stimulator was abolished. Five milliliters of the anesthetic solution were injected and after a 60 second waiting period the remainder of the anesthetic was injected. At the end of the administration, the patient complained of difficulty breathing.

The patient was alert and oriented, but he complained of dyspnea and short respiratory incursion. On inspection, breathing was predominantly intercostal ipsilateral to the blockade, with retraction of the abdominal wall and tachypnea. On auscultation breath sounds were absent in the right base. The left hemithorax maintained the same pre-operative pattern. A hypothetic diagnosis of ipsilateral phrenic nerve block was made.

The patient remained conscious with oxygen via nasal cannula in a semisitting position with the dorsum at 30° to reduce the respiratory discomfort. SpO2 remained at 95%. The dyspnea did not worsen and, therefore, non-invasive respiratory support (CPAP or BIPAP) or controlled mechanical ventilation was not necessary. A chest X-ray revealed the right hemidiaphragm at the level of the 6th rib, occupying the right 5th intercostal space (Figure 1).



The symptom eventually improved, the blockade was established and the surgery was performed without complaints of pain, discomfort or the need of supplemental analgesia. The surgery lasted approximately 75 minutes.

At the end of the procedure the patient was transferred to the post-anesthesia recovery unit where there was complete remission of the dyspnea after approximately two hours, with SpO2 96% in room air. Pulmonary auscultation revealed the same preoperative pattern, with breath sounds present in the right base. The brachial plexus block was maintained and the patient did not complain of pain. The patient was kept under observation in the recovery room for three hours before being referred to the admission unit. Five hours after the procedure the patient was admitted to the hospital and discharged in the following day.



In the brachial plexus block, the local anesthetic is injected inside the musculoaponeurotic cuff formed, proximally, by the anterior (aponeurosis of the posterior portion) and middle (aponeurosis of the anterior portion) scalene muscles 2,6.

The brachial plexus is formed by the ventral branches of the inferior cervical nerves, C5 to T1, with or without contributions from C4 and T2. The phrenic nerve originates from the deep cervical plexus, derived from the ventral branches of C2, C3, and C4, being located very close to the brachial plexus in the neck, in front of the anterior scalene muscle, separated from the plexus only by a this fascia 3. Therefore, it can be reached by the diffusing anesthetic solution when the perivascular interscalene and supraclavicular approaches are used; this is not seen with the perivascular axillary techniques due to the distance between the site of injection and the phrenic nerve. Some authors admit that almost all adverse events resulting from phrenic nerve block are due to extravasation of the local anesthetic out of the musculoaponeurotic cuff or its dispersion towards the cervical plexus, which, in the cuff, is contiguous to the brachial plexus 3,4. Thus, if the solution reaches the level between C3 and C5 it spurs the blockade. Not even digital compression above the site of the puncture prevent extravasation of the local anesthetic 7,8.

Usually, brachial plexus block is done by perivascular techniques because the musculoaponeurotic cuff that surrounds it maintains a direct relationship with arteries and veins in its entire path. In the classical description of the techniques, correct needle placement was confirmed by the development of paresthesia in the extremity to be blocked. It is known that this can be associated with lesion of the nerve by the needle and, therefore, confirmation of the correct needle positioning is done by carefully palpating the interscalene cleft and the use of the peripheral nerve stimulator. However, it should be noticed that the peripheral nerve stimulator does not prevent phrenic nerve block or even the intraneural injection of the local anesthetic 9. However, the sudden onset of diaphragmatic contractions during the neural stimulation indicates that the needle should be repositioned by moving it posteriorly, closer to the middle scalene muscle. A case of definitive phrenic nerve paralysis secondary to interscalene block, in which the peripheral nerve stimulator was used 9, has been reported. Ventilatory changes were compatible with restrictive pulmonary disease. Tidal volume, forced expiratory volume in 1 second, forced vital capacity, and total lung capacity were reduced to 98%, 79%, 88% and 76%, respectively. On the other hand, peak flow, PaO2, and PaCO2 remained within normal limits. The authors believe that the local anesthetic was injected intraneurally 9.

A study compared the diaphragmatic incursion in patients with physical status ASA I or II undergoing interscalene and axillary brachial plexus block 3. Diaphragmatic movements were evaluated by ultrasound on the ipsilateral and contralateral hemithoraces. Thirty-four to 52 mL of 1.5% mepivacaine with epinephrine were used for the blockade; the volume depended on the clinical demands of each case. Ipsilateral phrenic nerve block was observed in all patients with the interscalene approach, but without clinical repercussions. The axillary approach was not associated with any cases of phrenic nerve block. Paralysis reversed within three to five hours, despite the presence of anesthesia in the upper limb 3. Therefore, the duration of the blockade is determined by the pharmacological properties of the local anesthetic used and it is unlikely to cause severe compromise in healthy patients.

Symptoms of phrenic nerve block can vary from absence to relatively severe symptoms, depending on the presence of pre-existing conditions, especially pulmonary dysfunction or obesity 3,4.

A study with 30 patients without pulmonary disease evaluated the diaphragmatic incursion after perivascular supraclavicular brachial plexus block 4. A 75% or greater reduction in diaphragmatic movement was considered diaphragmatic paralysis, partial paralysis with a 25% to 75% reduction, and a reduction of less than 25% as absence of paralysis. Diaphragmatic movement was evaluated by ultrasound 10 minutes after the blockade. In the absence of paralysis, the patient was evaluated again 10 minutes after the first evaluation. Fifty percent of the patients presented complete diaphragmatic paralysis after the blockade and all of them presented complete blockade of the upper limb. Seventeen percent of the patients presented partial paralysis. Patients with diaphragmatic paralysis had a significant reduction in pulmonary function both in the sitting position and supine. Those with partial paralysis also presented a reduction in pulmonary function, but this difference was not statistically significant. SpO2 did not change, even when patients were breathing room air. They concluded that phrenic nerve blockade secondary to brachial plexus block is an important adverse reaction, which might affect adversely the pulmonary function, although it is well-tolerated by patients without pre-existing pulmonary disorders 4.

Phrenic nerve blockade also occurs in 40% to 70% 10 of the cases of perivascular subclavian brachial plexus block, as well as with the perivascular infraclavicular approach, leading to diaphragmatic paresis or even paralysis causing dyspnea and respiratory failure in susceptible patients.

In reality, most healthy patients tolerate ipsilateral diaphragmatic paralysis without any symptoms. It is possible that the accessory musculature compensates the restriction imposed by the paralysis and that expansion of the contralateral lung is able to produce enough negative pressure to guarantee good ventilation.

Development of dyspnea after brachial plexus block demands that other causes, such as pneumothorax, recurrent laryngeal nerve block, bronchospasm, allergic reaction, direct neurological lesion and injection in the neuro axis be ruled out. Phrenic nerve block can contribute to trigger the symptoms, respiratory effort and anxiety causing an increase in negative pressure in the upper airways but might not be the only cause of dyspnea.

The development of dyspnea is the result of the association of phrenic nerve block with other factors and despite studies that use ultrasound to evaluate diaphragmatic incursion inspiratory chest X-ray is a simple and useful exam to detect the problem and might be faster if it is available in the operating room. Ultrasound has the advantage of being able to determine the degree of diaphragmatic incursion and therefore demonstrate whether the patient has partial or total paralysis 3. Preoperative sedation with benzodiazepines does not seem to be related with the development of intraoperative dyspnea since in most cases patients seem to be cooperative and alert. Chronic obstructive pulmonary disease (COPD), such as emphysema, is an important co-factor for the development of symptoms. With the destruction of the pulmonary parenchyma, the diaphragmatic movement is important to guarantee the hematosis since a 50% loss in diaphragmatic function will result in dyspnea. In patients with a restrictive pulmonary pattern, the loss of diaphragmatic movements further impairs ventilation and might cause respiratory failure.

Therefore, in patients with COPD, overweight, of compromised respiratory function, the perivascular interscalene, supraclavicular, subclavian or infraclavicular approaches should have restricted indications. In those cases, one should if possible consider the perivascular axillary technique which is not associated with phrenic nerve block 4.

Since it is believed that phrenic nerve block is secondary to extravasation of the local anesthetic at the injection site, if the volume injected in the musculoaponeurotic cuff is reduced, it is possible to decrease the incidence of diaphragmatic paresis or paralysis. However, some authors do not agree with this assertion and one should not forget that a reduction in volume might decrease the efficacy of the blockade 3. A radiologic study on the dispersion of different volumes (20, 30, and 40 mL) in the posterior brachial plexus block demonstrated the dispersion to be similar to that of the interscalene approach. It also demonstrated clinical similarities and that, invariably, the cervical plexus is involved with the possibility of phrenic nerve block 11.

The case presented here demonstrated, unequivocally, the development of total phrenic nerve block in a patient with pulmonary compromise, which led to the development of the symptoms reported. However, even after reversal of the symptoms and discharge from the recovery room the patient remained under observation for three more hours in the post-anesthesia recovery unit, where the anesthesiologist and nursing staff are close to the patient and alert to treat immediately any intercurrence.

We presented a case of phrenic nerve paralysis associated with the perivascular interscalene approach in a patient with chronic renal failure with pulmonary compromise, who developed respiratory symptoms. Despite the good evolution and lack of the need to institute invasive treatment, this should alert anesthesiologists to restrict the indication of the technique in similar cases.



01. Vieira JL – Bloqueio do Plexo Braquial, em: Cangiani LM – Anestesia Ambulatorial, Rio de Janeiro, Atheneu, 2001;281-300.        [ Links ]

02. Vieira JL – Bloqueio do Plexo Braquial, em: Cangiani LM, Posso IP, Potério GMB et al. – Tratado de Anestesiologia SAESP, 6ª Ed, Rio de Janeiro, Atheneu, 2006;1255-1274.        [ Links ]

03. Urmey WF, Talts KH, Sharrock NE – One hundred percent incidence of hemidiaphragmatic paresis associated with interscalene brachial plexus anesthesia as diagnosed by ultrasonography. Anesth Analg, 1991;72:498-503.        [ Links ]

04. Mak PH, Irwin MG, Ooi CG et al. – Incidence of diaphragmatic paralysis following supraclavicular brachial plexus block and its effect on pulmonary function. Anaesthesia, 2001;56:352-356.        [ Links ]

05. Wilson JL, Brown DL, Wong GY et al. – Infraclavicular brachial plexus block, parasagittal anatomy important to the coracoid technique. Anesth Analg, 1998;87:870-873.        [ Links ]

06. Ward ME – The interscalene approach to the brachial plexus. Anaesthesia, 1974;29:147-157.        [ Links ]

07. Neal JM, Moore JM, Kopacz DJ et al. – Quantitative analysis of respiratory, motor, and sensory function after supraclavicular block. Anesth Analg, 1998;86:1239-1244.        [ Links ]

08. Knoblanche GE – The incidence and aetiology of phrenic nerve blockade associated with supraclavicular brachial plexus block. Anaesth Intensive Care, 1979;7:346-349.        [ Links ]

09. Bashein G, Robertson HT, Kennedy WF – Persistent phrenic nerve paresis following interscalene brachial plexus block. Anesthesiology, 1985;63:102-105.        [ Links ]

10. Hickey R, Ramamurthy S – The diagnosis of phrenic nerve block on chest x-ray by a double-exposure technique. Anesthesiology, 1989;70:704-707.        [ Links ]

11. Cruvinel MGC, Castro CHV, Silva YP et al. – Estudo radiológico da dispersão de diferentes volumes de anestésico local no bloqueio de plexo braquial pela via posterior. Rev Bras Anestesiol, 2005;55:508-516.        [ Links ]

Correspondence to
Dr. Luís Henrique Cangiani
Av. Princesa D'Oeste, 1.300/101
13100-040 Campinas, SP

Submitted em 26 de janeiro de 2007
Accepted para publicação em 23 de dezembro de 2007



* Recebido do Received from Centro Médico de Campinas, Hospital sede do CET/SBA em Anestesiologia do Instituto Penido Burnier, Campinas, SP

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License