The liver produces the vitamin K-dependant clotting factors II, VII, IX, and X, so that either liver dysfunction or vitamin K deficiency can reduce production of these factors. The INR is a good measure of the activity of these liver-derived clotting factors. Warfarin and other coumarins are competitive inhibitors of vitamin K production in the liver. Liver disease reduces factors V and VIII as well, and defective fibrinogen is often produced. Some reversal is common with administration of vitamin K parenterally, even in the presence of moderately severe liver failure.

Bleeding may arise due to low numbers of platelets or platelet dysfunction.

Platelets <100,000/mL are associated with an increased risk of bleeding, and platelet transfusion of 4–6 units of fresh platelets is usually required to increase platelet numbers and reduce this risk. Heparin-induced thrombocytopenia (HIT) is not uncommonly associated with heparin therapy, especially unfractionated heparin. Platelets are consumed relative rapidly so transfusion is usually preferable on the operating table, just prior to surgical incision.

Platelets may be normal in number, but unable to aggregate, typically due to ingestion of antiplatelet agents, such as aspirin (which is irreversible) or nonsteroidal anti-inflammatory agents (e.g., piroxicam, diclofenac, ibuprofen, indomethacin), which are reversible within a few days of cessation and clearance.

Transfusion of whole blood or separated blood components is often used where the hemoglobin has fallen below 7 g/L or rapid blood loss has occurred and urgent replacement is needed or where specific replacement of blood components is required.

Blood products include:

The risks of allogeneic transfusion are detailed with vascular surgery in Volume 6.

Hematological problems often further complicate multisystem organ failure (MSOF) arising from either drug therapy (e.g., heparin-induced thrombocytopenia; HIT), blood product transfusion reactions, disseminated intravascular coagulation (DIC), or clotting factor abnormalities (e.g., due to liver failure). Intractable coagulopathic bleeding is a common sequel of severe MSOF and is often difficult to reverse and a major contributing factor to death. DVT and/or pulmonary embolism are other complications of prolonged immobility and coagulopathies.

These are principally directed against development of deep venous thrombosis and pulmonary thromboembolism (DVT and PTE).

Measures to reduce risk of deep venous thrombosis (DVT) fall into three broad groups:

The effectiveness can vary dependent on the clinical procedure and health system or specific patient-related risk factors.

VTE is a spectrum of disease, from asymptomatic distal DVT to fatal PE, and most DVT (and PE) are initially asymptomatic. The clinical relevance of asymptomatic distal DVT as an indicator of the efficacy of VTE prophylaxis is debated. Confirmation of the clinical diagnosis of DVT is usually by compression duplex ultrasonography, to show venous anatomy and flow patterns. The “gold standard” diagnostic tool used in most historical clinical trials has been ascending venography, considered to be of greater sensitivity than compression ultrasound for distal (below-knee) DVT. Venography is an invasive technique, injecting radiographic contrast agents into the foot vein(s) and viewing the venous drainage patterns using serial radiographs. This has essentially been replaced by compression duplex ultrasonography in clinical practice in most western countries, as the ultrasound technology has advanced in reliability and become readily available. Radionucleotide methods of detection of DVT have been used in many trials but are often considered oversensitive and are overall less reliable and more invasive, with radiation risks. There is also some risk of inducing DVT from venography.

PE is usually diagnosed or excluded by computed tomographic (CT) pulmonary angiography (helical CT), ventilation-perfusion isotope radionucleotide scan, or occasionally magnetic resonance imaging (MRI). PE incidence in randomized controlled trials is probably underestimated because diagnosis was on clinical suspicion, followed by investigation, rather than routine scanning.

Between clinical trials, the patient groups, randomization methods, diagnosis, and therapies vary considerably and make comparison and interpretation potentially problematic and imprecise. The time periods over which these studies were performed also vary, with different patient risk profiles, hospitalization periods, mobilization schedules, perioperative management, anesthetic, and surgical types applying during these studies. How these results then apply to contemporary practice remains uncertain. Nevertheless, VTE, DVT, and PE remain as serious complications of hospitalization. Many trials have used preoperative thromboprophylaxis, but current practice in many institutions is to initiate the dosing just after surgery or on the next day. Therefore, significant lack of available evidence in important areas necessarily limits the scope of any evidence-based interpretation or recommendations. However, the relative risks from VTE (death, pulmonary dysfunction, pain) and thromboprophylaxis (bleeding, platelet dysfunction) need to be thoughtfully balanced for each patient. Despite this, the absolute risk of VTE is still not removed entirely even with diligent thromboprophylaxis. In addition, clinicians and patients may perceive the risk and consequences of bleeding significantly differently, with surgeons being very concerned about intraoperative bleeding and operative difficulty, while patients are often primarily concerned about their total operative course. An extremely good evidence-based resource is the National Health and Medical Research Council of Australia (2009) Clinical Practice Guideline document addressing Prevention of Venous Thromboembolism in Patients Admitted to Australian Hospitals.

Bleeding is a side effect of trauma and surgery from vessel laceration and injury. Hematoma and bruising are consequences of bleeding into tissues or spaces. The risk of bleeding is multifactorial and is increased where particularly vascular regions or organs are operated on, with extensive surgery and where coagulation is impaired. Because pharmacological thromboprophylaxis is a means of altering coagulation to reduce risk of clot formation, especially DVT, there is often an increased risk of operative or postoperative bleeding. This risk, of course, needs to be balanced and interpreted against the risk of VTE and may be contraindicated in certain situations.

Relative or absolute contraindications to pharmacological VTE prophylaxis are as follows¹:

This is a process where small airways close down principally due to under-breathing with loss of volume of alveolae and eventual collapse of alveolar units. This typically occurs at the lung bases and can often be reversed with good analgesia to permit more effective deep breathing and coughing. Lung expansion opens the airways, expands the alveoli, and restores perfusion to the re-ventilated alveolar units. A common early clinical sign of atelectasis is a tachycardia, usually without a fever, occurring in the early postoperative period. Good pain relief is essential for good lung expansion to avoid under-breathing, permit adequate deep breathing and coughing, and to encourage proper expansion of alveolar lung units to avoid atelectasis.

“Good pain relief is essential for good lung expansion to avoid under-breathing and permit adequate deep breathing and coughing.”

Once atelectasis is established, if material is then aspirated into the airways or if primary upper airways infection occurs, then established lung infection and inflammation can prevail. The patient usually develops a fever as an early clinical sign.

Where alveolar units are not being ventilated, oxygenation and clearance of carbon dioxide are reduced, so that perfusion of these units is inefficient and does not assist respiration. In this situation, a state of relative ventilation-perfusion (V:Q) inequality exists. In order to compensate for this inequality, blood is reflexly redirected away from non-ventilated alveolar units by selective vasoconstriction in the pulmonary vascular bed. This encourages more blood to reach those alveolar units and regions of the lung that are better ventilated than those that are poorly ventilated. This process more effectively matches ventilation with perfusion, enabling more effective oxygenation and CO₂ removal.

Where pneumonia is more widespread, involving a pulmonary lobe or lobes (or indeed a whole lung), the extent of V:Q inequality may be so severe as to produce profound hypoxemia. Because CO₂ diffuses much faster than O₂ and is breathed-off more rapidly than the O₂ can be absorbed, CO₂ does not usually accumulate in the blood as fast as O₂ levels decrease. Thus, hypoxemia typically occurs before hypercapnia develops.

The risk of atelectasis and pneumonia is associated with a variety of risk factors, which are listed in Table 2.11.

A number of risk reduction strategies are useful in patients at higher risk of lung infections, where possible: prior to surgery, during surgery, and after surgery (Table 2.12).

These measures represent key factors in reducing the risk of postoperative chest infection.

When airways become occluded either by foreign bodies or secretions, absorption of gases distal to the obstructing point can lead to collapse of alveoli and small airways. The affected section of lung can collapse partially or entirely. This may occur independently or as part of a pneumonic process. Clearance of the airway obstruction is paramount to expansion of the collapsed lung, either by surgical bronchoscopic removal of the foreign body or secretions and/or from antibiotic therapy and chest physiotherapy. The presence of a pleural effusion can be severely detrimental by reducing lung expansion, facilitating lung collapse, and creating poor mucus clearance and sometimes considerable discomfort. Drainage of a large pleural effusion is often essential for proper lung expansion, oxygenation, and avoidance or resolution of pneumonic consolidation.

Pneumothorax may arise from pulmonary barotrauma due to ventilation, direct surgical injury, chest tube trauma, or spontaneously, the latter often from apical bullous disease.

Pneumothorax is recognized as loss of chest expansion and air entry on the affected side and may be signified by chest pain and breathlessness in the alert patient, by or poor gas exchange.

Two main types occur:

Tension pneumothorax is due to progressive buildup of pressure from air leakage into the pleural space without release, as a “flap-valve” effect. The pressure typically displaces the mediastinum towards the opposite side, compressing the heart, great vessels in the chest, and opposite lung, leading to difficulty with respiration or ventilation, worsening gas exchange, and compromised venous return. This is a surgical emergency and needs to be corrected urgently by releasing the pressure in the chest by inserting a large-bore needle through the chest wall on the side of the pneumothorax.

Non–tension pneumothorax is where leakage of gas has occurred around the lung into the pleural space leading to collapse of the lung away from the chest wall; the pressure does not continue to build within the chest and so does not displace mediastinal structures nor cause cardiovascular compromise. Occasionally, a non-tension pneumothorax can evolve into a tension pneumothorax, especially when positive pressure ventilation is used after a (non-tension) pneumothorax has developed.

When gas leaks from the lung, pleural space, or airways into the tissues, the tissues can swell and develop crepitus (crackling) to palpation. In these cases, gas may often be seen tracking along between tissue planes on plain x-ray or CT scans. Occasionally this phenomenon can occur after laparotomy, but is not uncommon after laparoscopy or thoracoscopy, where CO₂ is insufflated. It is usually benign and resolves spontaneously over a few days. Surgical emphysema is an important sign to note and may be a serious concern for the patient if it is palpable, audible, extensive, or associated with a fever. Rarely, it can signify gas gangrene, superficial infections, rupture of the esophagus, an anastomotic leak, or result from peritoneal gas following viscus perforation; in these situations early diagnosis and treatment is paramount.

The lower airways and lung are usually virtually sterile containing few, if any, bacteria. When foreign material reaches the larynx or enters the airways, in an alert patient it typically causes extreme irritation and activation of the coughing reflex. Aspiration of food, other material, or regurgitated gastric contents can occur in the awake patient with an active gag reflex, but the airways are usually protected by reflex coughing and laryngeal spasm. In the sedated or unconscious patient, or in those with neurological problems, the risk of aspiration of material into the airways is much higher, due principally to depression of the protective reflex mechanisms.

Aspiration pneumonitis is the term usually given to inhalation of regurgitated gastric contents into the airways. Gastric contents are extremely irritant, and inflammation occurs rapidly, leading to bronchospasm with obstruction of smaller affected airways. An initial chemical pneumonitis occurs and may subsequently proceed to an infective pneumonia. Typical early chest x-ray appearances can be unremarkable; with progression, a focal right middle or lower lobe pneumonic picture may be seen or a patchy multifocal appearance evident. The left lung may be affected if the patient is lying on their left side.

Risk factors for aspiration pneumonitis:

Edema of the neck tissues for any reason may cause respiratory compromise either through direct pressure on the airways causing closure or from laryngeal spasm or both. Bleeding into the neck tissues (postoperative or otherwise) and neck trauma are the most common causes; however, allergic reactions with profound neck swelling, gross bloating from hypoproteinemia, burns involving the upper respiratory tract, and even severe surgical emphysema can produce breathing difficulties. Anaphylaxis is a rare event in most patients, but some patients are particularly prone, and it is typically associated with laryngospasm and neck swelling, in addition to bronchospasm.

Perhaps the commonest cause of tracheal or bronchial obstruction intraoperatively is from a misplaced endotracheal tube with obstruction of a main stem bronchus. Postoperatively, mucus plugging of the airway with thick inspissated secretions is most commonly observed.

Inhaled foreign bodies are the commonest preoperative cause in children and adults, and aspiration of solid gastric contents in trauma patients is not uncommon. Inhalation of a dislodged tooth either before or during laryngoscopy and intubation for anesthesia is well reported, although uncommon. Bronchoscopy may be required to remove the foreign material.

As mentioned previously, blood pressure is determined to three main factors:

Changes in blood pressure can be evaluated by consideration of these three factors in turn, and the cause(s) determined as being due to one or a combination of these.

If under-perfusion is significant enough, a state of “shock” may result causing hypoxemic organ injury or death.

Cardiac function and output is determined by heart rate and stroke volume. Stroke volume, in turn, is dependent on wall integrity, wall size, cardiac valve function, and cardiac muscle contractile function. It is often useful to analyze the potential cause(s) of cardiac failure by considering these factors.

Heart rate is a function of the electrical rhythmicity of the sinoatrial (SA) node which typically fires at 70–100 beats per minute, being conducted to the atrioventricular (AV) node and conducting system to cause cardiac muscle contractions. Disturbances in the rate or rhythm can cause abnormal filling of the heart and alter cardiac output. Sinus rhythm is regular in nature and varies with respiration due principally to differences in intrathoracic pressure and the amount of venous return, and thereby right heart filling. Heart rate is increased by sympathetic stimulation and decreased by parasympathetic stimulation. Drugs can influence heart rate by primarily mimicking these pathways and their receptors, for example, atropine can quicken (by removing vagal inhibitory tone) and beta-blockers may slow (by removing noradrenergic stimulatory effects) cardiac rate. If the SA node fails as a pacemaker, the next most active part of the atrium or ventricle takes over as the pacemaker; arrhythmias may ensue, for example, atrial fibrillation (AF), heart block, ventricular ectopic beats, or ventricular fibrillation. During myocardial ischemia and infarction, abnormal rhythm circuits can occur which can effectively take over cardiac rhythm and cause bizarre ventricular rhythms, most of which result in less effective cardiac output than normal.

Arrhythmias arise when normal sinus rhythm is disturbed by alteration of function of the normal pattern of conduction. These principally occur when the sinoatrial node fails to function (e.g., sinus arrest, sinus block, sinus delay, sinus arrhythmia) or another part of the atrium (or ventricle) takes over the pacemaker function. Other causes include septal infarction, which can involve the AV node or septal conducting system, or ventricular circus movements where part(s) of the ventricle causes cardiac contraction overriding the normal pathway and conduction pattern. Serious arrhythmias are the most common cardiac events responsible for major morbidity and mortality associated with surgery and the perioperative period. However, as discussed later, arrhythmias are often the consequence of myocardial ischemia or cardiac failure, with each entity continuously feeding back on the other to worsen the situation (i.e., more arrhythmia, causing more ischemia, causing more arrhythmia), unless the deleterious “cycle” can be broken by correcting the arrhythmia or ischemia or both. Common recognized arrhythmias are listed below in Table 2.13.

Stroke volume is influenced by factors presented in Table 2.14.

These factors are of course not mutually exclusive, most affect each other in some way, for example, ventricular hypertrophy influences both wall size and contractility.

These are the most common cardiac complications associated with the perioperative period. Infarction just prior to surgery substantially worsens the risk of mortality. The risk of perioperative mortality progressively reduces over a 3-month period after a myocardial infarction, then stabilizes; if the surgery can be delayed for 3 months, then the surgery-associated risk might be decreased. This is, of course, a highly patient- and procedure-dependent matter for individual consideration. The indication for surgery, on balance, may predicate early intervention. In some cases, coronary revascularization may be more important to perform before elective surgery. A history of myocardial ischemia, previous infarction, cardiac failure, or previous vascular compromise or surgery should alert the anesthetist and surgeon to the increased risk of a perioperative cardiac event. However, ischemia may be asymptomatic and therefore not evident before surgery, so a careful history is useful to detect any hint of possible cardiac disease. Optimization of the patient with known cardiac disease is discussed in Chap.​ 4.

The integrity of the blood vessels is paramount for containing the blood within a closed system to maintain blood pressure. Blood vessels may dilate and constrict in response to sympathetic and parasympathetic nervous system, physiological, and pathological factors. Peripheral vasodilatation can occur in shock, especially septic shock, where the vascular system expands in its capacity to cause a fall in blood pressure, although the cardiac function is often hyperdynamic and the blood volume may have increased in an attempt to compensate. Vasoconstriction may also occur in response to pain, hypothermia, injury, and cardiogenic shock. Some of these responses are “physiological,” although in the setting in which they occur they become “pathological,” in terms of the survival advantage to the patient. A normal response to hemorrhage would be generalized vasoconstriction of the gut and limbs, to redirect blood to the brain, lungs, and heart (physiological). However, when vascular shutdown is too intense and for too long in duration, organs may become ischemic and infarction may occur (pathological).

Blood volume is relative to the capacity of the vascular system as pointed out above. An important factor implicit in this consideration is the time frame over which any change in blood volume or vascular system capacity occurs. Rapid bleeding can result in a serious acute decline in blood volume such that vasoconstriction alone may be insufficient to compensate. Heart rate and stroke volume will usually rise to increase cardiac output, but this may not fully compensate for the loss of blood volume. Judicious and appropriate blood, blood product, or other fluid replacement may correct the blood volume deficit sufficiently to normalize blood pressure and organ perfusion. Slower bleeding or oozing can go undetected for some time, especially if clinically concealed, and may only become evident when clinical anemia develops or if acute bleeding supervenes. The time course of slower bleeding may permit adequate compensation of blood volume loss through vasoconstriction and by absorption of fluid from the gut to replace losses.

Although hypotension is perhaps the predominant complication that complicates surgical perioperative care, hypertension is not uncommonly encountered. Hypertension is usually silent and asymptomatic until a major cerebral or cardiac event supervenes, such as a hemorrhage or an infarct. Uncontrolled hypertension should be managed urgently to avoid these complications. Serial monitoring over 1 or more days may be appropriate to establish that persistent elevation is present, as opposed to a sporadic isolated reading. Intermittent high blood pressure may rarely herald the presence of a pheochromocytoma, which should be diagnosed and managed pharmacologically prior to any general anesthesia and surgery to avoid a malignant hypertensive crisis. With the exception of the anxiety of hospitalization for surgery, the commonest cause of hypertension in the perioperative period is failure of the patient to take their usual antihypertensive medications due to fasting, which when recommenced usually corrects the problem. Most anesthetists and physicians recommend taking the usual antihypertensives with a sip of fluid by the patient, well prior to surgery.

The Cochrane Collaboration Review (2007, updated 2009) concluded that there was no evidence that the risk of death was reduced using colloids vs. crystalloids, after resuscitation in patients requiring volume replacement, in patients with trauma, burns, or following surgery. Data was derived from 55 RCTs, which presented mortality data. The use of colloids was more expensive than crystalloids, and since the efficacy was comparable, use of colloids outside the context of RCTs was difficult to justify. Pregnant women and neonates were excluded.

Albumin or Plasma Protein Fraction – Twenty-three trials with 7,754 patients reported data on mortality. The relative risk (RR) from these trials was 1.01 (95 % confidence interval [95 % CI] 0.92 to 1.10). When a trial with poor quality allocation concealment was excluded, pooled RR was 1.00 (95 % CI 0.91 to 1.09).

Hydroxyethyl Starch – Sixteen trials with 637 patients compared hydroxyethyl starch with crystalloids. The RR was 1.05 (95 % CI 0.63 to 1.75).

Modified Gelatin – Eleven trials with 506 patients compared modified gelatin with crystalloid. The RR was 0.91 (95 % CI 0.49 to 1.72).

Dextran – Nine trials with 834 patients compared dextran with a crystalloid. The RR was 1.24 (95 % CI 0.94 to 1.65).

Colloids in Hypertonic Crystalloid Compared to Isotonic Crystalloid – Eight trials with 1,283 patients compared dextran in hypertonic crystalloid with isotonic crystalloid including. The RR was 0.88 (95 % CI 0.74 to 1.05).

Side effects of gels may be unanticipated and occasionally severe with development of allergic reactions, including anaphylaxis. These problems may complicate an already problematic clinical situation, which can be averted to some degree by very careful consideration of when and in whom to use these products.

Fluids used for replacement include:

Normal intake of fluid per day is of the order of 1.5 to 2 L per day, assuming fluid losses are normal and stable. Higher fluid losses may occur in certain situations such as high temperature environments where dehydration occurs and in disease states such as diarrhea and vomiting. After surgery, “third-space” fluid losses can be appreciable (up to 2–4 L) as fluid collects in swollen tissues as a result of the surgical tissue trauma. These situations often demand more fluid than the usual 2 L/day.

The required fluid input can be determined by urine output, provided renal function is normal, so that the amount of fluid infused per hour can usually be titrated against the urine output over the previous hour(s).

The amount of fluid and rate of infusion that needs to be used for resuscitation are determined by clinical judgement, relating to factors that include known or predicted blood or fluid loss, blood pressure level, degree of trauma or shock, urine output, central venous pressure, known comorbidities, and the age of the patient.

The main complications of fluid infusion are:

Overload is caused by infusion of too much fluid such that the vascular capacity is exceeded, with respect to the ability of the heart to adequately pump the fluid to clear the peripheral circulation (resulting in peripheral edema) and/or the ability to clear the lungs (resulting in pulmonary edema). These two aspects essentially reflect the functional capacity of the right and left heart to pump blood, respectively.

Underhydration is the failure to adequately replace fluid or maintain the sufficient fluid volume to keep up with fluid losses, to ensure adequate cardiac filling, circulation, and tissue perfusion. Both of these may precede surgery and complicate anesthesia and perioperative care.

In the intensive care or surgical high-dependency settings, and occasionally in the surgical wards, central venous pressure monitoring can provide a very useful tool to evaluate the amount of venous filling and circulatory load and to enable balancing of fluids more accurately.

In general terms, when nutritional intake cannot be restored by the usual oral feeding route within 7–10 days after surgery, a state of severe catabolism has commenced and malnutrition will usually result, unless some alternative form of nutritional maintenance is provided, whether by enteral or parenteral means.

The most effective form of nutrition is via the gut with nutrient absorption in the usual manner. This means that enteral nutrition provides the most preferable and predictable form of administering nutrition to patients, where at all possible.

Forms of enteral feeding include:

Enteral nutrition may be used preoperatively, where possible, to improve nutritional status as much as possible before surgery and to reduce the risks and complications from malnutrition after surgery. This may reverse any catabolic effects existing before surgery, depending on the clinical problem, to place the patient in a more favorable situation to be able to cope with the catabolic state that usually ensues, for a variable period, after surgery.

If a period of total parenteral nutrition (TPN) is required, either before and/or after surgery, then it is wise to move as soon as practicable to enteral forms of nutrition to provide better and more cost-efficient administration of nutrition.

Complications and problems associated with enteral feeding:

When enteral administration of nutrition is not possible, or inadvisable, then total parenteral nutrition (TPN) is able to be used to provide an intravenous form of balanced nutritional support. TPN requires administration of high osmolarity concentrated fluids via a large diameter vein, usually using a central venous catheter, because TPN fluids often require relatively rapid infusion and, due to their concentration, are extremely irritant to smaller peripheral veins. TPN is used where gut function is not possible or the gut route needs to be rested, either following gastrointestinal or other surgery, for example, where fistulae are trying to be closed or ischemic gut is present. The use of preoperative TPN represents a “two-edged sword” to some degree, because there is the benefit of reversal of the catabolic state with improvement of nutrition and immune function, but there is also a complication rate (up to 10 %) associated with central venous line insertion, maintenance, and TPN administration over time.

Sometimes, placement of a percutaneous gastroenterostomy tube or feeding enterostomy tube is useful during surgery to permit rapid transfer to enteral feeding after a period of TPN.

Components usually used in TPN include:

In many institutions, these are mixed to order in the hospital pharmacy under sterile conditions and delivered in bag(s) to the surgical high dependency, ICU, or ward as needed. Between bags of TPN, IV administration of 10 % glucose is advisable, as this will usually reduce the risk of hypoglycemia from the late effects of insulin contained in the TPN. When TPN is being ceased, then a scaling-down process, with gradual reduction of the rate of TPN administration is also a useful technique to avoid rapid and potentially dangerous changes in glucose or other levels in the blood.

Lipids are usually administered as 10, 20, or 30 % concentration emulsions of long-chain fatty acids, containing essential fatty acids (linoleic and linolenic acid).

Complications and problems associated with TPN:

Hypokalemia, hypophosphatemia, hypomagnesemia, and glucose intolerance are not uncommonly experienced when a severely malnourished patient is fed using TPN – the so-called refeeding syndrome.